{"gene":"GJD2","run_date":"2026-04-28T18:06:53","timeline":{"discoveries":[{"year":1998,"finding":"Cx36 (encoded by GJD2) was cloned as a new connexin family member of 321 amino acids, and in situ hybridization showed it is expressed predominantly in mammalian neurons (inferior olive, olfactory bulb, hippocampus, brainstem nuclei, retinal ganglion cell and inner nuclear layers), making it the first connexin shown to be expressed preferentially in neurons rather than glia.","method":"Degenerate RT-PCR cloning, genomic library isolation, in situ hybridization, neurotoxin lesion experiments","journal":"The European journal of neuroscience","confidence":"High","confidence_rationale":"Tier 1 — original cloning with multiple orthogonal methods and neurotoxin validation, highly cited foundational paper","pmids":["9753189"],"is_preprint":false},{"year":1999,"finding":"The human GJD2 gene was cloned and localized to chromosome 15q14; its coding sequence is 98% identical at the protein level to mouse/rat Cx36, and the gene structure (single intron 71 bp after the translation initiation site) is conserved across species, confirming it is the human ortholog of rodent Cx36.","method":"Gene cloning, sequencing, fluorescence in situ hybridization (FISH), radioactive in situ hybridization","journal":"Journal of neuroscience research","confidence":"High","confidence_rationale":"Tier 1-2 — direct cloning and chromosomal localization with multiple methods","pmids":["10462698"],"is_preprint":false},{"year":2000,"finding":"Cx36 is selectively expressed by insulin-producing beta-cells in the central region of pancreatic islets, with little or no expression in other endocrine cell types (alpha, delta cells), establishing beta-cell-specific gap junctional coupling via Cx36.","method":"In situ hybridization, immunolabeling with affinity-purified antibodies, FACS-purified beta- vs. non-beta-cell fractions","journal":"Diabetes","confidence":"High","confidence_rationale":"Tier 2 — multiple orthogonal methods, replicated across subsequent studies","pmids":["10905480"],"is_preprint":false},{"year":2001,"finding":"Freeze-fracture replica immunogold labeling (FRIL) demonstrated that Cx36 is exclusively localized to neuronal gap junctions in adult rat CNS, whereas Cx26, Cx30, and Cx43 are restricted to astrocytes and Cx32 to oligodendrocytes, establishing separate and non-overlapping connexin-defined pathways for neuronal versus glial gap junctional communication.","method":"Freeze-fracture replica immunogold labeling (FRIL), confocal immunocytochemistry","journal":"Cell communication & adhesion","confidence":"High","confidence_rationale":"Tier 1 — ultrastructural immunogold co-localization in >4000 labeled gap junctions in >370 replicas","pmids":["12064610"],"is_preprint":false},{"year":2001,"finding":"Global ischemia induced a selective increase in Cx36 protein (without corresponding mRNA increase) in parvalbumin-positive inhibitory interneurons of CA1 hippocampus before neuronal death, suggesting post-translational regulation of Cx36 and a role for Cx36 gap junctions in the survival of GABAergic interneurons.","method":"Western blot, Northern blot, in situ hybridization, double immunofluorescence, Cx32 knockout mouse model","journal":"The Journal of neuroscience","confidence":"Medium","confidence_rationale":"Tier 2 — multiple methods in one lab, but mechanistic link to survival is inferred from KO phenotype","pmids":["11567043"],"is_preprint":false},{"year":2002,"finding":"N-terminal domain mutagenesis of Cx36 identified a structural element necessary for normal subcellular localization; site-directed mutagenesis of putative phosphorylation motifs did not alter localization, indicating that phosphorylation/dephosphorylation is not the major regulatory step in Cx36 protein transport. Cx36-EGFP formed functional gap junction plaques with electrical properties indistinguishable from native Cx36.","method":"Cx36-EGFP transfection in neuroblastoma cell lines and primary hippocampal neurons, mutagenesis, electrophysiology","journal":"Journal of neuroscience research","confidence":"Medium","confidence_rationale":"Tier 2 — mutagenesis combined with functional electrophysiology and imaging","pmids":["12210839"],"is_preprint":false},{"year":2003,"finding":"The neuron-restrictive silencer factor (NRSF/REST) acts as a transcriptional repressor of the GJD2 (Cx36) gene via a conserved neuron-restrictive silencer element (NRSE) in the promoter; ectopic NRSF expression in beta-cell lines markedly reduced Cx36 mRNA and protein, and mutation of the NRSE relieved repression, explaining why Cx36 is expressed specifically in neurons and beta-cells (which lack NRSF).","method":"Luciferase reporter assays, viral gene transfer of NRSF into beta-cell lines, NRSE mutagenesis, trichostatin A (HDAC inhibitor) treatment","journal":"The Journal of biological chemistry","confidence":"High","confidence_rationale":"Tier 1-2 — promoter dissection with mutagenesis, gain-of-function, and pharmacological validation across multiple cell lines","pmids":["14565956"],"is_preprint":false},{"year":2003,"finding":"Cx36-dependent gap junction coupling controls cytosolic Ca2+ handling in populations of insulin-producing cells; loss of Cx36 expression altered Ca2+ transients, identifying a molecular link between Cx36 and the stimulus-secretion pathway for insulin secretion.","method":"Antisense Cx36 transfection in MIN6 cells, Ca2+ imaging, insulin secretion assays","journal":"Cell communication & adhesion","confidence":"Medium","confidence_rationale":"Tier 2 — functional Ca2+ imaging linked to Cx36 loss-of-function with secretion readout","pmids":["14681053"],"is_preprint":false},{"year":2004,"finding":"Mefloquine potently and selectively blocks Cx36 gap junction channels (IC50 ~300 nM) and Cx50 channels (IC50 ~1.1 µM) while only affecting Cx43, Cx32, and Cx26 at 10–100-fold higher concentrations; mefloquine (25 µM) blocked gap junctional coupling between neocortical interneurons in brain slices with minimal nonspecific effects on synaptic potentials or intrinsic properties.","method":"Whole-cell patch clamp in transfected N2A cells, acute neocortical brain slices, pharmacological dose-response","journal":"Proceedings of the National Academy of Sciences of the United States of America","confidence":"High","confidence_rationale":"Tier 1-2 — in vitro electrophysiology with dose-response and in situ slice validation; replicated in subsequent studies","pmids":["15297615"],"is_preprint":false},{"year":2004,"finding":"Cx36, but not E-cadherin, is specifically required for normal insulin secretion in MIN6 beta-cells; lentivirus-mediated restoration of Cx36 (but not E-cadherin) rescued insulin secretory defects in Cx36 antisense-transfected clones, demonstrating that Cx36 gap junction function specifically controls beta-cell secretion.","method":"Stable antisense transfection, lentiviral transduction, insulin secretion assays","journal":"Experimental cell research","confidence":"High","confidence_rationale":"Tier 2 — rescue experiment with specific connexin vs. control protein, multiple secretagogues tested","pmids":["15023528"],"is_preprint":false},{"year":2004,"finding":"A lacZ reporter driven by the Cx36 locus revealed expression in GABAergic neurons of cerebellar nuclei, non-GABAergic neurons of the inferior olive, mitral cells of the olfactory bulb, parvalbumin-positive cells of cerebral cortex, retinal layers, insulin-producing beta-cells of the pancreas, and the adrenal medulla, establishing the full tissue distribution of Cx36 in adult mice.","method":"Beta-galactosidase reporter gene expression in Cx36(+/del[LacZ]) transgenic mice, histochemistry","journal":"The Journal of comparative neurology","confidence":"High","confidence_rationale":"Tier 2 — genetic reporter strategy provides definitive cell-type attribution; highly cited","pmids":["15116387"],"is_preprint":false},{"year":2005,"finding":"Human and mouse microglia express Cx36 mRNA and protein, and Cx36 gap junctions form between microglia and neurons in co-culture, providing electrophysiological coupling below 30 pS consistent with Cx36 channel properties; this establishes a direct physical channel of communication between microglia and neurons via Cx36.","method":"RT-PCR, Western blot, immunofluorescence, electrophysiology (coupling measurements), Lucifer yellow dye coupling","journal":"Journal of neuroscience research","confidence":"Medium","confidence_rationale":"Tier 2 — multiple methods from single lab including electrophysiology","pmids":["16211561"],"is_preprint":false},{"year":2006,"finding":"FRIL ultrastructural analysis confirmed that Cx36 (and less abundantly Cx45) is present at neuronal gap junctions specifically at 'mixed' glutamatergic/electrical synapses between mitral/tufted cell dendrites in the olfactory bulb; Cx36 was absent from olfactory receptor neuron axons, and genomic analysis revealed multiple miRNA binding sites in the 3'-UTR of Cx36 consistent with post-transcriptional regulation.","method":"Freeze-fracture replica immunogold labeling (FRIL), confocal immunofluorescence, genomic miRNA binding site analysis","journal":"Journal of neurocytology","confidence":"High","confidence_rationale":"Tier 1 — ultrastructural FRIL co-localization at defined synapse types","pmids":["16841170"],"is_preprint":false},{"year":2007,"finding":"Cx36 gap junctional coupling electrically hyperpolarizes beta-cells by propagating the membrane potential of adjacent inactive cells; in Cx36 knockout mice with blocked KATP channels, beta-cells rapidly depolarized and showed continuous electrical activity rather than the hyperpolarized state seen in coupled cells. Loss of Cx36 also increased Ca2+ channel density and altered the kinetics of insulin secretion in response to glucose changes.","method":"Electrophysiology in pancreatic slices from Cx36-/-, Cx36+/-, and Cx36+/+ mice; Ca2+ channel recordings; comparison with NMRI mice","journal":"Diabetes","confidence":"High","confidence_rationale":"Tier 1-2 — direct electrophysiology in genetic knockout model with multiple functional readouts","pmids":["17395748"],"is_preprint":false},{"year":2007,"finding":"Cx36 channels in beta-cells preferentially permit exchange of cationic molecules (cations and positively charged tracers) over anionic molecules, demonstrating strong cation selectivity; glibenclamide stimulation increased coupling for all permeant molecules, while coupling extent varied depending on molecular charge.","method":"Microinjection of tracers differing in size and charge into islet cells of Cx36 knockout, transgenic, and wild-type mice; MIN6 cell coupling assays","journal":"Diabetologia","confidence":"High","confidence_rationale":"Tier 2 — systematic permeability study using multiple tracers across genetic models","pmids":["17828386"],"is_preprint":false},{"year":2008,"finding":"CaMKII directly binds to two juxtamembrane cytoplasmic domains of Cx36 and phosphorylates Cx36 in vitro; both binding domains resemble the pseudosubstrate and pseudotarget sites of CaMKII and show phosphorylation-dependent interaction and autonomous CaMKII activation, analogous to CaMKII interaction with the NR2B subunit of NMDA receptors. CaMKII and Cx36 were significantly co-localized in the inferior olive.","method":"35S-labeled CaMKII binding assay to Cx36 cytoplasmic domains, in vitro phosphorylation assay, colocalization by immunofluorescence in inferior olive","journal":"Proceedings of the National Academy of Sciences of the United States of America","confidence":"High","confidence_rationale":"Tier 1 — in vitro binding and phosphorylation with domain mapping; mechanistic analogy to NMDA receptor validated by multiple methods","pmids":["19095792"],"is_preprint":false},{"year":2008,"finding":"Human beta-cells express Cx36 protein at the membrane in lipid raft domains where it forms functional gap junctions that preferentially exchange cationic molecules; Cx36 mRNA levels correlate with insulin gene expression in both control and type 2 diabetic islets, establishing Cx36 as the native coupling protein of human beta-cells that contributes to gene expression regulation.","method":"Immunostaining, lipid raft fractionation, dye coupling assays, correlation analysis of Cx36 and insulin mRNA in human islets from T2D and control donors","journal":"Human molecular genetics","confidence":"High","confidence_rationale":"Tier 2 — lipid raft localization plus functional coupling in human tissue; multiple methods","pmids":["19000992"],"is_preprint":false},{"year":2009,"finding":"Cx36 directly interacts with the PDZ1 domain of ZO-2 and ZO-3 via its C-terminal SAYV PDZ-binding motif; co-IP from Cx36-transfected HeLa cells and betaTC-3 cells confirmed co-immunoprecipitation, and in vitro pull-down with truncated Cx36 lacking SAYV abolished binding, establishing ZO-2 and ZO-3 as scaffold partners at Cx36-containing gap junctions.","method":"Co-immunoprecipitation, in vitro PDZ domain pull-down assay, C-terminal peptide competition, truncation mutagenesis","journal":"Neurochemistry international","confidence":"High","confidence_rationale":"Tier 1-2 — reciprocal co-IP plus in vitro pull-down with domain and motif mutagenesis","pmids":["19418635"],"is_preprint":false},{"year":2012,"finding":"The transcription factor Beta2/NeuroD1 binds to three E-boxes in a 2-kbp fragment of the GJD2 promoter and, together with its cofactor E47, transactivates the promoter; this establishes Cx36 as a direct transcriptional target of Beta2/NeuroD1 during prenatal beta-cell differentiation, and Cx36 protein is selectively expressed by beta-cells throughout prenatal pancreas development.","method":"Reporter gene assay with GJD2 promoter fragments, ChIP-like binding assays, lentiviral expression of Beta2/NeuroD1, mouse pancreas developmental staging by immunolabeling","journal":"The Journal of membrane biology","confidence":"High","confidence_rationale":"Tier 1-2 — promoter dissection with E-box binding confirmation and gain-of-function transactivation assay","pmids":["22729650"],"is_preprint":false},{"year":2012,"finding":"Cx36 gap junction channels exhibit ~10-fold higher single-channel permeability (Pγ) for cationic dyes vs. anionic dyes of similar mass; Pγ for anionic dye Alexa-350 was 358-fold lower than Cx43, 138-fold lower than Cx40, confirming that Cx36 is one of the most cation-selective connexins, consistent with its role in K+-mediated electrical coupling.","method":"Dual whole-cell voltage clamp combined with dual-mode fluorescence imaging of single gap junction channels in HeLa cells stably expressing Cx36-EGFP","journal":"The Journal of membrane biology","confidence":"High","confidence_rationale":"Tier 1 — single-channel permeability measurements with rigorous quantitative comparison across connexin isoforms","pmids":["22752717"],"is_preprint":false},{"year":2012,"finding":"Intercellular synchronization of Ca2+ oscillations in MIN6 cells is a reliable measure of Cx36-dependent coupling; cells with reduced Cx36 show decreased Ca2+ synchrony, glibenclamide promotes Cx36 coupling and increases synchrony, and quinine inhibits it. Drug screens using this assay identified compounds affecting Cx36 distribution and insulin content.","method":"Semi-automatic fluorimetric Ca2+ imaging in MIN6 cell populations, pharmacological modulation of Cx36","journal":"PloS one","confidence":"Medium","confidence_rationale":"Tier 2 — functional coupling assay validated against genetic and pharmacological Cx36 perturbations","pmids":["22848521"],"is_preprint":false},{"year":2014,"finding":"Genetic knockout of Cx36 in mice expressing ATP-insensitive KATP channels (neonatal diabetes model) restored near-normal glucose-stimulated Ca2+ activity and insulin secretion, and prevented hyperglycemia; demonstrating that Cx36-mediated coupling propagates the hyperpolarizing effect of overactive KATP channels across beta-cells, and that reducing coupling can compensate for this defect.","method":"Double-transgenic mouse model (Cx36 KO × ATP-insensitive KATP), glucose homeostasis measurements, islet Ca2+ imaging, insulin secretion assays","journal":"Diabetes","confidence":"High","confidence_rationale":"Tier 2 — genetic epistasis experiment (double KO/transgenic) with multiple functional readouts","pmids":["24458355"],"is_preprint":false},{"year":2015,"finding":"Phosphorylation of Cx36 at Ser293 is significantly elevated in AII amacrine cells of rd10 and rd1 retinal degeneration mouse models compared to wildtype, using a phospho-specific antibody; this elevated phosphorylation state is associated with increased gap junction coupling that underlies aberrant spontaneous hyperactivity in degenerating retina.","method":"Immunofluorescence with phospho-Ser293 specific antibody in rd10 and rd1 mouse retinas vs. wildtype","journal":"Frontiers in cellular neuroscience","confidence":"Medium","confidence_rationale":"Tier 3 — single lab, single method (phospho-specific antibody) with functional inference","pmids":["26483638"],"is_preprint":false},{"year":2016,"finding":"Calmodulin (CaM) binds to the carboxy-terminus of Cx36 in a calcium-dependent manner at a site overlapping the CaMKII binding site; NMR solution structure reveals CaM binds Cx36 in its characteristic compact state with major hydrophobic contacts at W277 (anchor position 1) and V284 (position 8); Ca2+-loaded CaM activates Cx36 channels, unlike its effect on other connexins.","method":"NMR spectroscopy, Ca2+-dependent binding assays, electrophysiology of Cx36 channel activity after CaM application","journal":"Frontiers in molecular neuroscience","confidence":"High","confidence_rationale":"Tier 1 — NMR structure of Cx36-CaM complex with functional validation of channel activation","pmids":["27917108"],"is_preprint":false},{"year":2016,"finding":"A synonymous SNP (rs3743123, S196S) in GJD2 alters gap junction plaque formation and cell coupling in HeLa cells transfected with the variant cDNA compared to wildtype; transgenic mice expressing the variant under insulin promoter showed postnatal reduction of islet Cx36 levels and beta-cell survival, causing hyperglycemia, demonstrating functional consequences of this silent polymorphism.","method":"Transfection of HeLa cells, transgenic mouse lines with insulin promoter-driven variant vs. WT Cx36, immunostaining, glucose measurements","journal":"PloS one","confidence":"Medium","confidence_rationale":"Tier 2 — cell and in vivo functional studies but single lab","pmids":["26959991"],"is_preprint":false},{"year":2017,"finding":"CaMKII-β and CaMKII-δ co-localize specifically with Cx36-containing gap junctions in the mouse retina; in the outer retina only CaMKII-β co-localizes with Cx36, while in the inner retina both CaMKII-β and -δ co-localize with Cx36, suggesting isoform-specific regulation of electrical synapses in different retinal layers.","method":"Confocal microscopy with double-labeling for Cx36 and four CaMKII isoforms in mouse retina","journal":"Frontiers in molecular neuroscience","confidence":"Medium","confidence_rationale":"Tier 3 — single lab, co-localization only, no direct functional interaction test","pmids":["29311815"],"is_preprint":false},{"year":2018,"finding":"E3 ubiquitin ligases LNX1 and LNX2 co-localize with Cx36-containing gap junctions in adult mouse brain, co-immunoprecipitate with Cx36, and LNX PDZ2 domain pulls down Cx36; cotransfection of E3 ligase-competent LNX isoforms with Cx36 caused loss of Cx36 gap junction plaques between cells, whereas ligase-inactive isoforms did not, establishing LNX-mediated ubiquitination as a mechanism for Cx36 internalization.","method":"Immunofluorescence colocalization, co-immunoprecipitation, PDZ pull-down, cotransfection of wild-type vs. ligase-dead LNX isoforms, LNX null mice","journal":"The European journal of neuroscience","confidence":"High","confidence_rationale":"Tier 2 — reciprocal co-IP plus pull-down plus functional ligase activity comparison in cells, with KO mouse controls","pmids":["30295974"],"is_preprint":false},{"year":2018,"finding":"Three unique cysteine residues (C264 in TM4, C92 and C87 in TM2) form a specific docking pocket in Cx36 accessible only to short-chain n-alcohols and isoflurane (which stimulate Cx36 GJ conductance) but not to inhibitory compounds; cysteine substitution mutations at these sites reversed the stimulatory effect of hexanol and isoflurane to inhibitory, suggesting that stimulation involves reshuffling of inter-subunit disulfide bonds (C264-C92) to intra-subunit bonds (C264-C87).","method":"Structural modeling/docking, site-directed mutagenesis of cysteines, dual whole-cell patch-clamp in HeLa and N2A cells","journal":"Bioscience reports","confidence":"Medium","confidence_rationale":"Tier 1-2 — mutagenesis with functional electrophysiology, but mechanistic model is partially computational","pmids":["29298877"],"is_preprint":false},{"year":2020,"finding":"A Cx36-GCaMP fusion biosensor formed functional gap junctions with PKA-regulated Neurobiotin coupling; local Ca2+ increases around Cx36 gap junctions were resolved in response to Ca2+ ionophore and glutamate application; glutamate potentiated Cx36-GCaMP coupling in HeLa cells via endogenous NMDA/kainate-type glutamate receptors and a Ca2+/CaMKII-dependent mechanism, demonstrating that glutamate receptor activation locally regulates Cx36 coupling strength through Ca2+ and CaMKII.","method":"Cx36-GCaMP fusion protein, tracer coupling assay, Ca2+ imaging, pharmacological CaMKII inhibition, RNA-seq confirmation of glutamate receptor expression","journal":"eNeuro","confidence":"Medium","confidence_rationale":"Tier 2 — novel biosensor approach with pharmacological validation; single lab","pmids":["32179580"],"is_preprint":false},{"year":2021,"finding":"Depletion of zebrafish gjd2a (Cx35.5, ortholog of mammalian GJD2) caused hyperopia and electrophysiological changes in the retina, while depletion of gjd2b (Cx35.1) caused nuclear cataract that triggered axial elongation; establishing that GJD2 orthologs regulate ocular biometry and refractive status through retinal and lenticular mechanisms.","method":"Zebrafish gjd2 morpholino/CRISPR knockouts, ocular biometry, electroretinography, immunohistochemistry, scRNA sequencing","journal":"Communications biology","confidence":"High","confidence_rationale":"Tier 2 — loss-of-function with multiple functional readouts in a vertebrate model, two paralogs distinguished","pmids":["34083742"],"is_preprint":false},{"year":2021,"finding":"Cx36 expression in zebrafish ortholog GJD2 and retinal gap junctions (particularly in the IPL) regulate emmetropization; form-deprivation decreased total Cx36 and phosphorylated Cx36 (P-Cx36) and the P-Cx36/Cx36 ratio in the IPL, while pharmacological uncoupling with 18-β-GA induced myopia and reduced Cx36 and P-Cx36 in both IPL and OPL, implicating Cx36 phosphorylation state as a functional readout in refractive development.","method":"Form-deprivation and lens-induced myopia in guinea pigs, subconjunctival injection of gap junction uncoupler 18-β-GA, quantitative immunofluorescence of Cx36 and P-Cx36, neurobiotin cut-loading","journal":"Investigative ophthalmology & visual science","confidence":"Medium","confidence_rationale":"Tier 2 — pharmacological and deprivation models with quantitative phospho-Cx36 readout; single lab","pmids":["34283211"],"is_preprint":false},{"year":2021,"finding":"CaM and CaMKII binding to Cx36 are both calcium-dependent, with overlapping binding sites on the Cx36 C-terminus; CaM can engage Cx36 outside the gap junction plaque. The review synthesizes evidence that these Ca2+-dependent interactions constitute a core regulatory mechanism for Cx36 electrical synapse plasticity: Ca2+ elevation leads to CaM binding, which activates CaMKII, which phosphorylates Cx36 to modulate coupling strength.","method":"Review synthesizing binding assays, NMR structure, electrophysiology, and phosphorylation data from multiple studies","journal":"International journal of molecular sciences","confidence":"Medium","confidence_rationale":"Tier 2 — mechanistic model supported by primary data from multiple prior studies","pmids":["33922931"],"is_preprint":false},{"year":2023,"finding":"Cryo-EM structures of human Cx36 gap junction channel at 2.2–3.6 Å resolution revealed a dynamic equilibrium between closed and open states: in the closed state, channel pores are obstructed by lipids with N-terminal helices (NTHs) excluded from the pore; in the open state, pore-lining NTHs create a more acidic pore than Cx26 or Cx46/50, explaining Cx36's strong cation selectivity. Channel opening involves an α-to-π-helix transition in TM1 that weakens protomer-protomer interactions.","method":"Cryo-electron microscopy at 2.2–3.6 Å resolution, structural analysis of open and closed states","journal":"Nature communications","confidence":"High","confidence_rationale":"Tier 1 — near-atomic resolution cryo-EM structure providing mechanistic basis for cation selectivity and gating","pmids":["36906653"],"is_preprint":false},{"year":2024,"finding":"Cryo-EM structures of human Cx36 in complex with inhibitors (mefloquine, arachidonic acid, 1-hexanol) showed that all three competitively bind a pocket at the N-terminal helices (NTH), inducing a conformational shift from the pore-lining NTH (PLN) state to a flexible NTH (FN) state that allows lipids to obstruct the channel pore, revealing the molecular mechanism of Cx36 channel inhibition.","method":"Cryo-electron microscopy structures with inhibitor-bound Cx36, structural comparison of PLN vs. FN conformational states","journal":"Nature communications","confidence":"High","confidence_rationale":"Tier 1 — cryo-EM structures with three distinct inhibitors revealing a shared competitive binding mechanism","pmids":["39455592"],"is_preprint":false},{"year":2024,"finding":"The C-terminal tip of Cx36 mediates two distinct interactions in the early secretory pathway: the terminal valine acts as an ER export signal recruiting COPII cargo receptors Sec24A/B/C at ER exit sites, while the PDZ-binding motif SAYV mediates interaction with Golgi-stacking protein Grasp55 to stabilize Cx36 in the Golgi. Sec24 promotes ER exit, while Grasp55 retains Cx36 in the Golgi, establishing opposing regulatory roles in Cx36 trafficking.","method":"HEK293T expression system, siRNA knockdown, BioID proximity labeling screens, co-immunoprecipitation, confocal microscopy of ER vs. Golgi localization, overexpression studies","journal":"Cellular and molecular life sciences","confidence":"High","confidence_rationale":"Tier 2 — BioID screens combined with functional siRNA and co-IP validation of two distinct trafficking interactions","pmids":["39395036"],"is_preprint":false},{"year":2025,"finding":"Hemichannel permeability assays showed that Cx36 hemichannels selectively allow ATP release but not glutamate or lactate; mutational analysis of N-terminus charge and N-terminus–TM2 interactions (informed by differential selectivity between Cx46 and Cx50) identified these structural elements as key determinants of permselectivity, showing that molecular size alone cannot explain connexin hemichannel selectivity.","method":"Co-expression of genetically encoded fluorescent sensors for ATP, glutamate, and lactate with connexins in cells; N-terminus mutagenesis; pharmacological stimulation","journal":"bioRxiv (preprint)","confidence":"Medium","confidence_rationale":"Tier 1-2 — functional permeability assay with mutagenesis, but preprint only","pmids":["bio_10.1101_2025.03.12.642803"],"is_preprint":true},{"year":2025,"finding":"Electrostatic interactions (ESIs) at the second extracellular (E2) loop interface are required for Cx36 gap junction formation: at least three ESI residue pairs per E2-E2 interface must be intact for functional GJ conductance. These unique ESIs also ensure Cx36 docking specificity to itself, preventing heterotypic GJ formation with other brain connexins (Cx26, Cx30, Cx31.3, Cx32, Cx43, Cx45, Cx47).","method":"Computational electrostatic calculations, systematic missense mutagenesis of E2 interface residues, dual patch-clamp in HEK293 cell pairs","journal":"bioRxiv (preprint)","confidence":"Medium","confidence_rationale":"Tier 1-2 — structure-guided mutagenesis with electrophysiological validation, but preprint only","pmids":["bio_10.1101_2025.10.25.684567"],"is_preprint":true},{"year":2025,"finding":"ZO-1 interacts with Cx36 via PDZ1 domain binding to the Cx36 PDZ-binding motif (SAYV), and this interaction is inherently weak; molecular dynamics simulations and binding assays identified a single substitution at position 319 and acidic residues adjacent to the PBM that evolutionarily weaken this interaction. CaMKII-mediated phosphorylation of Cx36 reduces ZO-1 binding, suggesting that ZO-1 unbinding is a necessary event during potentiation of electrical synapses.","method":"Gaussian accelerated molecular dynamics, binding affinity assays, site-directed mutagenesis at PBM and adjacent residues, CaMKII phosphorylation assays","journal":"bioRxiv (preprint)","confidence":"Medium","confidence_rationale":"Tier 1-2 — computational + binding assay + phosphorylation data, but preprint only","pmids":["bio_10.1101_2025.10.27.684941"],"is_preprint":true},{"year":2025,"finding":"Cx36 gap junctions in the outer retina lower visual thresholds (i.e., increase sensitivity) under dim light, while inner retinal Cx36 gap junctions are required for maintaining thresholds; complete Cx36 KO increased visual thresholds 16.5-fold relative to controls, and outer retina-specific Cx36 removal lowered thresholds 2.6-fold, demonstrating layer-specific and opposing roles of Cx36 in setting visual detection thresholds via the rod OFF-pathway.","method":"Transgenic mice with retinal layer-specific Cx36 knockout, operant behavioral assay, psychophysical modeling","journal":"iScience","confidence":"High","confidence_rationale":"Tier 2 — layer-specific genetic knockout with quantitative behavioral readout and computational modeling","pmids":["41623460"],"is_preprint":false},{"year":2025,"finding":"In pancreatic islets, delta cell activation precedes beta cell Ca2+ oscillations at high glucose, and selective Cx36 knockout in delta cells confirmed that low-density Cx36 gap junctions contribute to delta-beta cell coordination; however, blockade of vesicle release (paracrine signaling) eliminated coupling between most delta and beta cells, demonstrating that beta-delta coordination is primarily driven by paracrine signals (Urocortin 3) with secondary contribution from Cx36 gap junctions.","method":"GCaMP6s Ca2+ imaging of hundreds of cells simultaneously, selective delta-cell Cx36 knockout, Rho-GTPase inhibitor ML-141 to block vesicle release","journal":"Proceedings of the National Academy of Sciences of the United States of America","confidence":"High","confidence_rationale":"Tier 2 — cell-type-specific KO combined with pharmacological dissection and large-scale Ca2+ imaging","pmids":["40956879"],"is_preprint":false},{"year":2024,"finding":"In vivo BioID proximity labeling using Cx36-EGFP in AII amacrine cells identified more than 50 new proteins associated with Cx36 electrical synapses in mouse retina, including scaffold proteins, adhesion molecules, and cytoskeletal regulators; ZO-1 and ZO-2 were confirmed, and Sipa1l3 was identified as a new electrical synapse scaffold that interacts with ZO-1, ZO-2, and Cx36.","method":"In vivo BioID proximity labeling (GFP-nanobody-TurboID in mice; Cx35b-TurboID transgenic zebrafish), mass spectrometry, immunofluorescence, binding interaction assays","journal":"bioRxiv (preprint)","confidence":"Medium","confidence_rationale":"Tier 2 — in vivo proximity labeling with cross-species validation and direct binding confirmation for Sipa1l3, but preprint only","pmids":["bio_10.1101_2024.11.26.625481"],"is_preprint":true}],"current_model":"GJD2-encoded Cx36 is the principal gap junction protein of neuronal electrical synapses and pancreatic beta-cells: it forms homotypic, strongly cation-selective gap junction channels whose open/closed states involve lipid occlusion and N-terminal helix repositioning (established by cryo-EM); channel conductance is potentiated by Ca²⁺-loaded calmodulin binding to the C-terminal tail (W277/V284 anchors) and by CaMKII phosphorylation of the same domain (analogous to NR2B regulation), while LNX1/LNX2 E3 ubiquitin ligases ubiquitinate Cx36 to drive its internalization; trafficking from ER to Golgi is controlled by C-terminal valine (COPII/Sec24 ER export signal) and the SAYV PDZ-binding motif (Grasp55 Golgi retention), with ZO-1/ZO-2/ZO-3 anchoring Cx36 at synaptic plaques via PDZ1 interactions that are weakened by CaMKII phosphorylation; in beta-cells, Cx36-dependent coupling synchronizes Ca²⁺ oscillations and insulin secretion, and its expression is transcriptionally controlled by Beta2/NeuroD1 (activator) and NRSF/REST (repressor); in the retina, Cx36 in distinct layers differentially controls visual thresholds and emmetropization."},"narrative":{"teleology":[{"year":1998,"claim":"The identity of a neuron-specific connexin was unknown; cloning of Cx36 from mammalian brain established GJD2 as the first connexin expressed preferentially in neurons rather than glia, defining a new molecular basis for neuronal electrical synapses.","evidence":"Degenerate RT-PCR cloning, in situ hybridization with neurotoxin lesion validation in rat brain","pmids":["9753189"],"confidence":"High","gaps":["Functional channel properties not yet determined","No loss-of-function data"]},{"year":2000,"claim":"Whether Cx36 operated outside the CNS was unclear; detection of Cx36 selectively in pancreatic beta-cells established a second major physiological context and linked gap junctions to insulin secretion biology.","evidence":"In situ hybridization, immunolabeling, FACS-purified beta- vs. non-beta-cell fractions in rodent islets","pmids":["10905480"],"confidence":"High","gaps":["Functional consequence for insulin secretion not yet demonstrated","Human beta-cell expression not confirmed at this point"]},{"year":2001,"claim":"Whether neuronal and glial gap junctions share connexins was debated; ultrastructural FRIL demonstrated that Cx36 is exclusively neuronal, with no overlap with astrocytic (Cx43/Cx30/Cx26) or oligodendrocytic (Cx32) connexins.","evidence":"Freeze-fracture replica immunogold labeling of >4000 gap junctions in adult rat CNS","pmids":["12064610"],"confidence":"High","gaps":["Regional variation in co-expression with other neuronal connexins (e.g. Cx45) not fully resolved"]},{"year":2003,"claim":"How Cx36 expression is restricted to neurons and beta-cells was unknown; identification of NRSF/REST as a transcriptional repressor acting via an NRSE element in the GJD2 promoter explained tissue specificity—cells lacking NRSF (neurons, beta-cells) express Cx36.","evidence":"Luciferase reporter assays, NRSF viral gain-of-function in beta-cell lines, NRSE mutagenesis, HDAC inhibitor treatment","pmids":["14565956"],"confidence":"High","gaps":["Other transcription factors (beyond NRSF) controlling Cx36 not yet identified","Chromatin-level regulation not mapped"]},{"year":2004,"claim":"Whether Cx36 was functionally required for beta-cell secretion (versus a bystander) was unresolved; lentiviral rescue of Cx36 (but not E-cadherin) in Cx36-depleted MIN6 cells demonstrated specific and non-redundant control of insulin secretion by Cx36 gap junctions.","evidence":"Stable antisense transfection, lentiviral rescue, insulin secretion assays with multiple secretagogues","pmids":["15023528"],"confidence":"High","gaps":["In vivo relevance in whole-animal glucose homeostasis not yet tested"]},{"year":2007,"claim":"The biophysical mechanism by which Cx36 controls beta-cell excitability was unclear; electrophysiology in Cx36 KO pancreatic slices showed that Cx36 propagates hyperpolarizing currents from inactive cells, dampening excitability across the islet, and that Cx36 channels are strongly cation-selective.","evidence":"Patch-clamp electrophysiology in pancreatic slices from Cx36-/-, +/-, +/+ mice; tracer permeability with multiple charged dyes","pmids":["17395748","17828386"],"confidence":"High","gaps":["Structural basis for cation selectivity not yet determined","Molecular identity of permeant species in vivo unknown"]},{"year":2008,"claim":"How Cx36 coupling strength is dynamically regulated was unknown; discovery that CaMKII directly binds and phosphorylates Cx36 at cytoplasmic domains analogous to the CaMKII–NR2B interaction established a Ca²⁺-dependent plasticity mechanism for electrical synapses.","evidence":"In vitro CaMKII binding and phosphorylation assays with Cx36 cytoplasmic peptides, immunofluorescence colocalization in inferior olive","pmids":["19095792"],"confidence":"High","gaps":["Phosphorylation sites not mapped to specific residues at this stage","In vivo consequence of phosphorylation on coupling not demonstrated"]},{"year":2009,"claim":"How Cx36 is anchored at gap junction plaques was unresolved; identification of ZO-2 and ZO-3 as PDZ1-domain partners binding the C-terminal SAYV motif established a scaffold complex at Cx36 electrical synapses.","evidence":"Reciprocal co-immunoprecipitation, in vitro PDZ pull-down, C-terminal truncation mutagenesis in HeLa and betaTC-3 cells","pmids":["19418635"],"confidence":"High","gaps":["Whether ZO-1 also binds Cx36 directly not confirmed until later studies","Functional consequence of scaffold disruption on coupling not tested"]},{"year":2012,"claim":"The developmental transcriptional activator of Cx36 in beta-cells was unknown; Beta2/NeuroD1 was shown to bind E-boxes in the GJD2 promoter and transactivate it, establishing Cx36 as a direct transcriptional target during prenatal beta-cell differentiation.","evidence":"Reporter gene assays with GJD2 promoter fragments, E-box binding assays, lentiviral NeuroD1 expression, developmental immunolabeling","pmids":["22729650"],"confidence":"High","gaps":["Whether NeuroD1 also controls Cx36 in neurons not tested","Epigenetic regulation of the GJD2 locus not addressed"]},{"year":2016,"claim":"How Ca²⁺ directly modulates Cx36 channels was structurally undefined; NMR resolution of the CaM–Cx36 C-terminal complex revealed Ca²⁺-loaded CaM binds at W277/V284 in a compact conformation and activates Cx36 channels, unlike its inhibitory role on other connexins.","evidence":"NMR solution structure, Ca²⁺-dependent binding assays, electrophysiological channel activation measurements","pmids":["27917108"],"confidence":"High","gaps":["Full-length Cx36–CaM complex structure not available","Whether CaM and CaMKII binding are mutually exclusive in situ remains unclear"]},{"year":2018,"claim":"How Cx36 gap junctions are turned over was unknown; LNX1/LNX2 E3 ubiquitin ligases were identified as direct interactors that ubiquitinate Cx36 and drive gap junction plaque internalization, establishing ubiquitin-dependent degradation as a removal mechanism.","evidence":"Co-IP, PDZ pull-down, cotransfection of ligase-active vs. ligase-dead LNX isoforms, LNX null mice","pmids":["30295974"],"confidence":"High","gaps":["Specific ubiquitination sites on Cx36 not mapped","Whether LNX acts at all Cx36-expressing tissues not tested"]},{"year":2023,"claim":"The structural basis for Cx36 gating and cation selectivity was unknown; cryo-EM structures at 2.2–3.6 Å revealed that channel opening requires an α-to-π-helix transition in TM1 that positions N-terminal helices into a negatively charged pore, while closure involves lipid occlusion of the channel.","evidence":"Cryo-electron microscopy of human Cx36 gap junction channels in open and closed conformations","pmids":["36906653"],"confidence":"High","gaps":["Dynamics of lipid entry/exit not resolved","No structures with CaM or CaMKII-phosphorylated Cx36"]},{"year":2024,"claim":"How pharmacological inhibitors block Cx36 was structurally undefined; cryo-EM structures with mefloquine, arachidonic acid, and hexanol showed all three bind a common N-terminal helix pocket, shifting it from a pore-lining to a flexible state that permits lipid occlusion, unifying the inhibition mechanism.","evidence":"Cryo-EM structures of inhibitor-bound Cx36 with conformational state analysis","pmids":["39455592"],"confidence":"High","gaps":["No activator-bound structures available","Whether this mechanism extends to in vivo neuromodulatory inhibition is untested"]},{"year":2024,"claim":"How Cx36 traffics from ER to plasma membrane was poorly understood; identification of the C-terminal valine as a COPII/Sec24 ER-export signal and the SAYV PDZ motif as a Grasp55-mediated Golgi retention signal established opposing regulatory checkpoints in the early secretory pathway.","evidence":"BioID proximity labeling, siRNA knockdown, co-immunoprecipitation, confocal ER/Golgi localization in HEK293T cells","pmids":["39395036"],"confidence":"High","gaps":["Whether these trafficking signals are regulated by phosphorylation in vivo is unknown","Post-Golgi trafficking steps not mapped"]},{"year":2025,"claim":"Whether Cx36 gap junctions serve identical functions across retinal layers was unclear; layer-specific knockout showed that outer retinal Cx36 lowers visual thresholds (increases sensitivity) while inner retinal Cx36 maintains thresholds, revealing opposing functional roles of the same connexin in different circuit contexts.","evidence":"Transgenic mice with retinal layer-specific Cx36 knockout, operant behavioral assay, psychophysical modeling","pmids":["41623460"],"confidence":"High","gaps":["Circuit-level mechanism by which inner vs. outer Cx36 differentially set thresholds not fully resolved","Contribution of Cx36 phosphorylation state in each layer not tested"]},{"year":2025,"claim":"The relative contribution of Cx36 gap junctions versus paracrine signaling to delta-beta cell coordination in islets was debated; selective delta-cell Cx36 KO and vesicle-release blockade demonstrated that paracrine signals (Urocortin 3) dominate, with Cx36 providing a secondary coordination mechanism.","evidence":"GCaMP6s Ca²⁺ imaging, delta-cell-specific Cx36 knockout, Rho-GTPase inhibitor to block vesicle release","pmids":["40956879"],"confidence":"High","gaps":["Whether Cx36 coupling between delta and beta cells is physiologically regulated remains unclear","Direct measurement of delta-beta electrical coupling not performed"]},{"year":null,"claim":"Key unresolved questions include: (1) the full-length structure of Cx36 in complex with CaM or in a CaMKII-phosphorylated state; (2) how post-Golgi trafficking delivers Cx36 to specific synaptic sites; (3) the identity and roles of most proteins in the Cx36 electrical synapse proteome identified by proximity labeling; and (4) whether Cx36 dysfunction directly causes human Mendelian disease.","evidence":"","pmids":[],"confidence":"Low","gaps":["No CaM-bound or phospho-Cx36 cryo-EM structure","Post-Golgi trafficking pathway undefined","Most BioID-identified interactors lack functional validation","No causative human mutations in GJD2 linked to Mendelian disease with direct evidence"]}],"mechanism_profile":{"molecular_activity":[{"term_id":"GO:0005215","term_label":"transporter activity","supporting_discovery_ids":[0,8,14,19,32,33]},{"term_id":"GO:0098631","term_label":"cell adhesion mediator activity","supporting_discovery_ids":[3,12,32]}],"localization":[{"term_id":"GO:0005886","term_label":"plasma membrane","supporting_discovery_ids":[3,12,16,32,33]},{"term_id":"GO:0005829","term_label":"cytosol","supporting_discovery_ids":[5]},{"term_id":"GO:0005794","term_label":"Golgi apparatus","supporting_discovery_ids":[34]},{"term_id":"GO:0005783","term_label":"endoplasmic reticulum","supporting_discovery_ids":[34]}],"pathway":[{"term_id":"R-HSA-1500931","term_label":"Cell-Cell communication","supporting_discovery_ids":[3,12,13,14,32,33]},{"term_id":"R-HSA-112316","term_label":"Neuronal System","supporting_discovery_ids":[0,3,8,12,15,38]},{"term_id":"R-HSA-162582","term_label":"Signal Transduction","supporting_discovery_ids":[7,13,15,20,21,28]},{"term_id":"R-HSA-9609507","term_label":"Protein localization","supporting_discovery_ids":[34]},{"term_id":"R-HSA-382551","term_label":"Transport of small molecules","supporting_discovery_ids":[14,19,32]}],"complexes":["Cx36 homotypic gap junction channel (dodecameric)"],"partners":["TJP1","TJP2","TJP3","CALM1","CAMK2B","LNX1","LNX2","GORASP2"],"other_free_text":[]},"mechanistic_narrative":"GJD2 encodes connexin 36 (Cx36), the principal gap junction protein of neuronal electrical synapses and pancreatic beta-cells, where it forms homotypic, strongly cation-selective intercellular channels that synchronize membrane potential and Ca²⁺ oscillations [PMID:9753189, PMID:10905480, PMID:22752717]. Cryo-EM structures reveal that channel gating involves an α-to-π-helix transition in TM1 that repositions N-terminal helices into the pore lumen, while channel closure is achieved by lipid occlusion when inhibitors (mefloquine, arachidonic acid) competitively displace N-terminal helices from their pore-lining configuration [PMID:36906653, PMID:39455592]. Channel conductance is potentiated by Ca²⁺/calmodulin binding at the C-terminal tail (W277/V284 anchors) and by CaMKII phosphorylation of overlapping cytoplasmic domains, while CaMKII phosphorylation simultaneously weakens ZO-1 scaffold interactions to remodel electrical synapse architecture; LNX1/LNX2 E3 ubiquitin ligases ubiquitinate Cx36 to drive gap junction plaque internalization [PMID:27917108, PMID:19095792, PMID:19418635, PMID:30295974]. In beta-cells, Cx36-dependent coupling propagates hyperpolarizing currents to coordinate insulin secretion, with transcription controlled by Beta2/NeuroD1 (activator) and NRSF/REST (repressor), while in the retina, layer-specific Cx36 gap junctions differentially set visual detection thresholds and regulate emmetropization [PMID:17395748, PMID:22729650, PMID:14565956, PMID:41623460, PMID:34083742]."},"prefetch_data":{"uniprot":{"accession":"Q9UKL4","full_name":"Gap junction delta-2 protein","aliases":["Connexin-36","Cx36","Gap junction alpha-9 protein"],"length_aa":321,"mass_kda":36.1,"function":"One gap junction consists of a cluster of closely packed pairs of transmembrane channels, the connexons, through which materials of low MW diffuse from one cell to a neighboring cell","subcellular_location":"Cell membrane; Cell junction, gap junction","url":"https://www.uniprot.org/uniprotkb/Q9UKL4/entry"},"depmap":{"release":"DepMap","has_data":true,"is_common_essential":false,"resolved_as":"","url":"https://depmap.org/portal/gene/GJD2","classification":"Not Classified","n_dependent_lines":0,"n_total_lines":1208,"dependency_fraction":0.0},"opencell":{"profiled":false,"resolved_as":"","ensg_id":"","cell_line_id":"","localizations":[],"interactors":[],"url":"https://opencell.sf.czbiohub.org/search/GJD2","total_profiled":1310},"omim":[{"mim_id":"607058","title":"GAP JUNCTION PROTEIN, DELTA-2; GJD2","url":"https://www.omim.org/entry/607058"},{"mim_id":"160700","title":"MYOPIA 2, AUTOSOMAL DOMINANT; MYP2","url":"https://www.omim.org/entry/160700"},{"mim_id":"121011","title":"GAP JUNCTION PROTEIN, BETA-2; GJB2","url":"https://www.omim.org/entry/121011"}],"hpa":{"profiled":true,"resolved_as":"","reliability":"","locations":[],"tissue_specificity":"Tissue enhanced","tissue_distribution":"Detected in some","driving_tissues":[{"tissue":"adrenal gland","ntpm":5.6},{"tissue":"pituitary gland","ntpm":7.3},{"tissue":"retina","ntpm":19.8}],"url":"https://www.proteinatlas.org/search/GJD2"},"hgnc":{"alias_symbol":["CX36"],"prev_symbol":["GJA9"]},"alphafold":{"accession":"Q9UKL4","domains":[{"cath_id":"1.20.1440.80","chopping":"15-109_191-293","consensus_level":"medium","plddt":87.4547,"start":15,"end":293}],"viewer_url":"https://alphafold.ebi.ac.uk/entry/Q9UKL4","model_url":"https://alphafold.ebi.ac.uk/files/AF-Q9UKL4-F1-model_v6.cif","pae_url":"https://alphafold.ebi.ac.uk/files/AF-Q9UKL4-F1-predicted_aligned_error_v6.png","plddt_mean":72.44},"mouse_models":{"mgi_url":"https://www.informatics.jax.org/marker/summary?nomen=GJD2","jax_strain_url":"https://www.jax.org/strain/search?query=GJD2"},"sequence":{"accession":"Q9UKL4","fasta_url":"https://rest.uniprot.org/uniprotkb/Q9UKL4.fasta","uniprot_url":"https://www.uniprot.org/uniprotkb/Q9UKL4/entry","alphafold_viewer_url":"https://alphafold.ebi.ac.uk/entry/Q9UKL4"}},"corpus_meta":[{"pmid":"9753189","id":"PMC_9753189","title":"Cloning of a new gap junction gene (Cx36) highly expressed in mammalian brain neurons.","date":"1998","source":"The European journal of neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/9753189","citation_count":377,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"15297615","id":"PMC_15297615","title":"Potent block of Cx36 and Cx50 gap junction channels by mefloquine.","date":"2004","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/15297615","citation_count":271,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"10751658","id":"PMC_10751658","title":"Expression of Cx36 in mammalian neurons.","date":"2000","source":"Brain research. Brain research reviews","url":"https://pubmed.ncbi.nlm.nih.gov/10751658","citation_count":216,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"12064610","id":"PMC_12064610","title":"Identification of cells expressing Cx43, Cx30, Cx26, Cx32 and Cx36 in gap junctions of rat brain and spinal cord.","date":"2001","source":"Cell communication & adhesion","url":"https://pubmed.ncbi.nlm.nih.gov/12064610","citation_count":179,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"17395748","id":"PMC_17395748","title":"Cx36-mediated coupling reduces beta-cell heterogeneity, confines the stimulating glucose concentration range, and affects insulin release kinetics.","date":"2007","source":"Diabetes","url":"https://pubmed.ncbi.nlm.nih.gov/17395748","citation_count":140,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"10905480","id":"PMC_10905480","title":"Cx36 preferentially connects beta-cells within pancreatic islets.","date":"2000","source":"Diabetes","url":"https://pubmed.ncbi.nlm.nih.gov/10905480","citation_count":136,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"11567043","id":"PMC_11567043","title":"Global ischemia-induced increases in the gap junctional proteins connexin 32 (Cx32) and Cx36 in hippocampus and enhanced vulnerability of Cx32 knock-out mice.","date":"2001","source":"The Journal of neuroscience : the official journal of the Society for Neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/11567043","citation_count":119,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"19000992","id":"PMC_19000992","title":"Cx36 makes channels coupling human pancreatic beta-cells, and correlates with insulin expression.","date":"2008","source":"Human molecular genetics","url":"https://pubmed.ncbi.nlm.nih.gov/19000992","citation_count":98,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"10462698","id":"PMC_10462698","title":"Structure, chromosomal localization, and brain expression of human Cx36 gene.","date":"1999","source":"Journal of neuroscience research","url":"https://pubmed.ncbi.nlm.nih.gov/10462698","citation_count":90,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"16841170","id":"PMC_16841170","title":"Ultrastructural localization of connexins (Cx36, Cx43, Cx45), glutamate receptors and aquaporin-4 in rodent olfactory mucosa, olfactory nerve and olfactory bulb.","date":"2006","source":"Journal of neurocytology","url":"https://pubmed.ncbi.nlm.nih.gov/16841170","citation_count":76,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"22569071","id":"PMC_22569071","title":"Impaired β-cell-β-cell coupling mediated by Cx36 gap junctions in prediabetic mice.","date":"2012","source":"American journal of physiology. Endocrinology and metabolism","url":"https://pubmed.ncbi.nlm.nih.gov/22569071","citation_count":61,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"11117498","id":"PMC_11117498","title":"Cx36 is dynamically expressed during early development of mouse brain and nervous system.","date":"2000","source":"Neuroreport","url":"https://pubmed.ncbi.nlm.nih.gov/11117498","citation_count":46,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"15023528","id":"PMC_15023528","title":"Differentiating the effects of Cx36 and E-cadherin for proper insulin secretion of MIN6 cells.","date":"2004","source":"Experimental cell research","url":"https://pubmed.ncbi.nlm.nih.gov/15023528","citation_count":42,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"12210839","id":"PMC_12210839","title":"Evidence for a role of the N-terminal domain in subcellular localization of the neuronal connexin36 (Cx36).","date":"2002","source":"Journal of neuroscience research","url":"https://pubmed.ncbi.nlm.nih.gov/12210839","citation_count":42,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"18337497","id":"PMC_18337497","title":"Zebrafish early cardiac connexin, Cx36.7/Ecx, regulates myofibril orientation and heart morphogenesis by establishing Nkx2.5 expression.","date":"2008","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/18337497","citation_count":26,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"29311815","id":"PMC_29311815","title":"Differential Distribution of Retinal Ca2+/Calmodulin-Dependent Kinase II (CaMKII) Isoforms Indicates CaMKII-β and -δ as Specific Elements of Electrical Synapses Made of Connexin36 (Cx36).","date":"2017","source":"Frontiers in molecular neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/29311815","citation_count":26,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"34083742","id":"PMC_34083742","title":"Loss of Gap Junction Delta-2 (GJD2) gene orthologs leads to refractive error in zebrafish.","date":"2021","source":"Communications biology","url":"https://pubmed.ncbi.nlm.nih.gov/34083742","citation_count":24,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"27917108","id":"PMC_27917108","title":"Structural and Functional Consequences of Connexin 36 (Cx36) Interaction with Calmodulin.","date":"2016","source":"Frontiers in molecular neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/27917108","citation_count":24,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"20709095","id":"PMC_20709095","title":"Visual responses in the lateral geniculate evoked by Cx36-independent rod pathways.","date":"2010","source":"Vision research","url":"https://pubmed.ncbi.nlm.nih.gov/20709095","citation_count":21,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"24458355","id":"PMC_24458355","title":"Decreasing cx36 gap junction coupling compensates for overactive KATP channels to restore insulin secretion and prevent hyperglycemia in a mouse model of neonatal diabetes.","date":"2014","source":"Diabetes","url":"https://pubmed.ncbi.nlm.nih.gov/24458355","citation_count":20,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"28841859","id":"PMC_28841859","title":"A GJA9 frameshift variant is associated with polyneuropathy in Leonberger dogs.","date":"2017","source":"BMC genomics","url":"https://pubmed.ncbi.nlm.nih.gov/28841859","citation_count":19,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"22897942","id":"PMC_22897942","title":"Connexin36 (Cx36) expression and protein detection in the mouse carotid body and myenteric plexus.","date":"2012","source":"Acta histochemica","url":"https://pubmed.ncbi.nlm.nih.gov/22897942","citation_count":18,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"26959991","id":"PMC_26959991","title":"A Variant of GJD2, Encoding for Connexin 36, Alters the Function of Insulin Producing β-Cells.","date":"2016","source":"PloS one","url":"https://pubmed.ncbi.nlm.nih.gov/26959991","citation_count":17,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"31776212","id":"PMC_31776212","title":"Rod Photoreceptors Signal Fast Changes in Daylight Levels Using a Cx36-Independent Retinal Pathway in Mouse.","date":"2019","source":"The Journal of neuroscience : the official journal of the Society for Neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/31776212","citation_count":17,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"14681053","id":"PMC_14681053","title":"Cx36 involvement in insulin secretion: characteristics and mechanism.","date":"2003","source":"Cell communication & adhesion","url":"https://pubmed.ncbi.nlm.nih.gov/14681053","citation_count":17,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"26483638","id":"PMC_26483638","title":"Increased phosphorylation of Cx36 gap junctions in the AII amacrine cells of RD retina.","date":"2015","source":"Frontiers in cellular neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/26483638","citation_count":17,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"28561933","id":"PMC_28561933","title":"Cx36, Cx43 and Cx45 in mouse and rat cerebellar cortex: species-specific expression, compensation in Cx36 null mice and co-localization in neurons vs. glia.","date":"2017","source":"The European journal of neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/28561933","citation_count":17,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"12064624","id":"PMC_12064624","title":"Cx36 and the function of endocrine pancreas.","date":"2001","source":"Cell communication & adhesion","url":"https://pubmed.ncbi.nlm.nih.gov/12064624","citation_count":17,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"22848521","id":"PMC_22848521","title":"The intercellular synchronization of Ca2+ oscillations evaluates Cx36-dependent coupling.","date":"2012","source":"PloS one","url":"https://pubmed.ncbi.nlm.nih.gov/22848521","citation_count":15,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"30016355","id":"PMC_30016355","title":"Neuroprotective effects of leonurine against oxygen-glucose deprivation by targeting Cx36/CaMKII in PC12 cells.","date":"2018","source":"PloS one","url":"https://pubmed.ncbi.nlm.nih.gov/30016355","citation_count":14,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"23834555","id":"PMC_23834555","title":"Evaluating the association between pathological myopia and SNPs in RASGRF1. ACTC1 and GJD2 genes at chromosome 15q14 and 15q25 in a Chinese population.","date":"2013","source":"Ophthalmic genetics","url":"https://pubmed.ncbi.nlm.nih.gov/23834555","citation_count":14,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"29476948","id":"PMC_29476948","title":"Cx36 in the mouse hippocampus during and after pilocarpine-induced status epilepticus.","date":"2018","source":"Epilepsy research","url":"https://pubmed.ncbi.nlm.nih.gov/29476948","citation_count":13,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"20599911","id":"PMC_20599911","title":"Role of Cx36 gap junction modulation in general anaesthetic anticonvulsant action.","date":"2010","source":"European journal of pharmacology","url":"https://pubmed.ncbi.nlm.nih.gov/20599911","citation_count":12,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"22729650","id":"PMC_22729650","title":"Cx36 is a target of Beta2/NeuroD1, which associates with prenatal differentiation of insulin-producing β cells.","date":"2012","source":"The Journal of membrane biology","url":"https://pubmed.ncbi.nlm.nih.gov/22729650","citation_count":12,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"35262731","id":"PMC_35262731","title":"The Role of GJD2(Cx36) in Refractive Error Development.","date":"2022","source":"Investigative ophthalmology & visual science","url":"https://pubmed.ncbi.nlm.nih.gov/35262731","citation_count":11,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"27462070","id":"PMC_27462070","title":"The role of Cx36 and Cx43 in 4-aminopyridine-induced rhythmic activity in the spinal nociceptive dorsal horn: an electrophysiological study in vitro.","date":"2016","source":"Physiological reports","url":"https://pubmed.ncbi.nlm.nih.gov/27462070","citation_count":11,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"31283648","id":"PMC_31283648","title":"Decreased expression of gap junction delta-2 (GJD2) messenger RNA and connexin 36 protein in form-deprivation myopia of guinea pigs.","date":"2019","source":"Chinese medical journal","url":"https://pubmed.ncbi.nlm.nih.gov/31283648","citation_count":11,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"34406332","id":"PMC_34406332","title":"Phenotypic Consequences of the GJD2 Risk Genotype in Myopia Development.","date":"2021","source":"Investigative ophthalmology & visual science","url":"https://pubmed.ncbi.nlm.nih.gov/34406332","citation_count":10,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"32179580","id":"PMC_32179580","title":"Localized Calcium Signaling and the Control of Coupling at Cx36 Gap Junctions.","date":"2020","source":"eNeuro","url":"https://pubmed.ncbi.nlm.nih.gov/32179580","citation_count":9,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"12363395","id":"PMC_12363395","title":"Mutational analysis of the connexin 36 gene (CX36) and exclusion of the coding sequence as a candidate region for catatonic schizophrenia in a large pedigree.","date":"2002","source":"Schizophrenia research","url":"https://pubmed.ncbi.nlm.nih.gov/12363395","citation_count":9,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"29208420","id":"PMC_29208420","title":"The inhibitor of connexin Cx36 channels, mefloquine, inhibits voltage-dependent Ca2+ channels and insulin secretion.","date":"2017","source":"Molecular and cellular endocrinology","url":"https://pubmed.ncbi.nlm.nih.gov/29208420","citation_count":9,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"33922931","id":"PMC_33922931","title":"The Roles of Calmodulin and CaMKII in Cx36 Plasticity.","date":"2021","source":"International journal of molecular sciences","url":"https://pubmed.ncbi.nlm.nih.gov/33922931","citation_count":8,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"34283211","id":"PMC_34283211","title":"The Role of Retinal Connexins Cx36 and Horizontal Cell Coupling in Emmetropization in Guinea Pigs.","date":"2021","source":"Investigative ophthalmology & visual science","url":"https://pubmed.ncbi.nlm.nih.gov/34283211","citation_count":7,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"28729916","id":"PMC_28729916","title":"Dynamic Expression of CX36 Protein in Kainic Acid Kindling induced Epilepsy.","date":"2017","source":"Translational neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/28729916","citation_count":6,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"29298877","id":"PMC_29298877","title":"Molecular basis for potentiation of Cx36 gap junction channel conductance by n-alcohols and general anesthetics.","date":"2018","source":"Bioscience reports","url":"https://pubmed.ncbi.nlm.nih.gov/29298877","citation_count":6,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"34827092","id":"PMC_34827092","title":"Extensive GJD2 Expression in the Song Motor Pathway Reveals the Extent of Electrical Synapses in the Songbird Brain.","date":"2021","source":"Biology","url":"https://pubmed.ncbi.nlm.nih.gov/34827092","citation_count":5,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"36126263","id":"PMC_36126263","title":"Astragaloside IV ameliorates cerebral ischemia-reperfusion injury via upregulation of PKA and Cx36.","date":"2022","source":"Neuroreport","url":"https://pubmed.ncbi.nlm.nih.gov/36126263","citation_count":4,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"31408166","id":"PMC_31408166","title":"Cadmium versus Lanthanum Effects on Spontaneous Electrical Activity and Expression of Connexin Isoforms Cx26, Cx36, and Cx45 in the Human Fetal Cortex.","date":"2020","source":"Cerebral cortex (New York, N.Y. : 1991)","url":"https://pubmed.ncbi.nlm.nih.gov/31408166","citation_count":4,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"40956879","id":"PMC_40956879","title":"Heterogeneity in the coordination of delta cells with beta cells is driven by both paracrine signals and low-density Cx36 gap junctions.","date":"2025","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/40956879","citation_count":3,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"33278187","id":"PMC_33278187","title":"Altered Expression of GJD2 Messenger RNA and the Coded Protein Connexin 36 in Negative Lens-induced Myopia of Guinea Pigs.","date":"2020","source":"Optometry and vision science : official publication of the American Academy of Optometry","url":"https://pubmed.ncbi.nlm.nih.gov/33278187","citation_count":3,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"31495909","id":"PMC_31495909","title":"Gap junctions and expression of Cx36, Cx43 and Cx45 in the posterodorsal medial amygdala of adult rats.","date":"2019","source":"Histology and histopathology","url":"https://pubmed.ncbi.nlm.nih.gov/31495909","citation_count":3,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"26084632","id":"PMC_26084632","title":"Cx36 expression in the AII-mediated rod pathway is activity dependent in the developing rabbit retina.","date":"2015","source":"Developmental neurobiology","url":"https://pubmed.ncbi.nlm.nih.gov/26084632","citation_count":2,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"35114378","id":"PMC_35114378","title":"Early decrease in Cx36 is associated with increased cell adhesion molecules (CAMs) junctional content in mouse pancreatic islets after short-term high-fat diet feeding.","date":"2022","source":"Annals of anatomy = Anatomischer Anzeiger : official organ of the Anatomische Gesellschaft","url":"https://pubmed.ncbi.nlm.nih.gov/35114378","citation_count":2,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"35885949","id":"PMC_35885949","title":"Association of CX36 Protein Encoding Gene GJD2 with Refractive Errors.","date":"2022","source":"Genes","url":"https://pubmed.ncbi.nlm.nih.gov/35885949","citation_count":1,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"38585986","id":"PMC_38585986","title":"Trafficking of Connexin36 (Cx36) in the early secretory pathway.","date":"2024","source":"bioRxiv : the preprint server for biology","url":"https://pubmed.ncbi.nlm.nih.gov/38585986","citation_count":1,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"39395036","id":"PMC_39395036","title":"Regulation of Cx36 trafficking through the early secretory pathway by COPII cargo receptors and Grasp55.","date":"2024","source":"Cellular and molecular life sciences : CMLS","url":"https://pubmed.ncbi.nlm.nih.gov/39395036","citation_count":1,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"39455592","id":"PMC_39455592","title":"Mefloquine-induced conformational shift in Cx36 N-terminal helix leading to channel closure mediated by lipid bilayer.","date":"2024","source":"Nature communications","url":"https://pubmed.ncbi.nlm.nih.gov/39455592","citation_count":1,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"39100594","id":"PMC_39100594","title":"Characterizing ER Retention Defects of PDZ Binding Deficient Cx36 Mutants Using Confocal Microscopy.","date":"2024","source":"Bio-protocol","url":"https://pubmed.ncbi.nlm.nih.gov/39100594","citation_count":1,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"38994720","id":"PMC_38994720","title":"Upregulation of the long noncoding RNA GJA9-MYCBP and PVT1 is a potential diagnostic biomarker for acute lymphoblastic leukemia.","date":"2024","source":"Cancer reports (Hoboken, N.J.)","url":"https://pubmed.ncbi.nlm.nih.gov/38994720","citation_count":0,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"39980061","id":"PMC_39980061","title":"Whole exome sequencing identified mutations of forkhead box I 1 (FOXI1), keratin 6 C (KRT6C) and gap junction protein delta 2 (GJD2) in a low-grade oncocytic tumor of the kidney: a case report.","date":"2025","source":"Diagnostic pathology","url":"https://pubmed.ncbi.nlm.nih.gov/39980061","citation_count":0,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"41623460","id":"PMC_41623460","title":"Retinal Cx36 gap junctions in the inner and outer plexiform layers differentially control visual thresholds.","date":"2025","source":"iScience","url":"https://pubmed.ncbi.nlm.nih.gov/41623460","citation_count":0,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"39828229","id":"PMC_39828229","title":"(-) - (11R, 12S)-mefloquine ameliorates neuropathic pain by modulating Cx36-ER stress interaction in the pain-related central nervous system in rats.","date":"2025","source":"Life sciences","url":"https://pubmed.ncbi.nlm.nih.gov/39828229","citation_count":0,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":"26692140","id":"PMC_26692140","title":"Characterization of the zebrafish cx36.7 gene promoter: Its regulation of cardiac-specific expression and skeletal muscle-specific repression.","date":"2015","source":"Gene","url":"https://pubmed.ncbi.nlm.nih.gov/26692140","citation_count":0,"is_preprint":false,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2025.10.25.684567","title":"Electrostatic Interaction Determines the Docking Specificity and the Formation of Functional Cx36 Gap Junction Channels","date":"2025-10-26","source":"bioRxiv","url":"https://doi.org/10.1101/2025.10.25.684567","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2025.10.27.684941","title":"Molecular determinants of low affinity complexes formed by the electrical synapse proteins Connexin 36 and ZO-1","date":"2025-10-27","source":"bioRxiv","url":"https://doi.org/10.1101/2025.10.27.684941","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2025.09.23.678084","title":"Connexin 36-mediated gap junctions contribute to fine odor discrimination and excitation of mitral cells in the mouse olfactory bulb","date":"2025-09-25","source":"bioRxiv","url":"https://doi.org/10.1101/2025.09.23.678084","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2024.11.26.625481","title":"Uncovering the electrical synapse proteome in retinal neurons via  <i>in vivo</i>  proximity labeling","date":"2024-11-26","source":"bioRxiv","url":"https://doi.org/10.1101/2024.11.26.625481","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2024.12.11.625486","title":"Dopamine regulates the membrane potential and glycine release of AII amacrine cells via D1-like receptor modulation of gap junction coupling","date":"2024-12-13","source":"bioRxiv","url":"https://doi.org/10.1101/2024.12.11.625486","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2024.11.20.624571","title":"Diminished gap junction coupling under diabetogenic conditions does not drive loss of functional β-cell sub-populations","date":"2024-11-22","source":"bioRxiv","url":"https://doi.org/10.1101/2024.11.20.624571","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2024.10.30.621116","title":"Rod inputs arrive at horizontal cell somas in mouse retina solely via rod-cone coupling","date":"2024-10-31","source":"bioRxiv","url":"https://doi.org/10.1101/2024.10.30.621116","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":null,"id":"bio_10.1101_2025.03.12.642803","title":"Mechanisms of permselectivity of connexin hemichannels to small molecules","date":"2025-03-12","source":"bioRxiv","url":"https://doi.org/10.1101/2025.03.12.642803","citation_count":0,"is_preprint":true,"source_track":"pubmed_title"},{"pmid":"12477932","id":"PMC_12477932","title":"Generation and initial analysis of more than 15,000 full-length human and mouse cDNA sequences.","date":"2002","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/12477932","citation_count":1479,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"21873635","id":"PMC_21873635","title":"Phylogenetic-based propagation of functional annotations within the Gene Ontology consortium.","date":"2011","source":"Briefings in bioinformatics","url":"https://pubmed.ncbi.nlm.nih.gov/21873635","citation_count":656,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"15489334","id":"PMC_15489334","title":"The status, quality, and expansion of the NIH full-length cDNA project: the Mammalian Gene Collection (MGC).","date":"2004","source":"Genome research","url":"https://pubmed.ncbi.nlm.nih.gov/15489334","citation_count":438,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"23396134","id":"PMC_23396134","title":"Genome-wide meta-analyses of multiancestry cohorts identify multiple new susceptibility loci for refractive error and myopia.","date":"2013","source":"Nature genetics","url":"https://pubmed.ncbi.nlm.nih.gov/23396134","citation_count":387,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"20835239","id":"PMC_20835239","title":"A genome-wide association study identifies a susceptibility locus for refractive errors and myopia at 15q14.","date":"2010","source":"Nature genetics","url":"https://pubmed.ncbi.nlm.nih.gov/20835239","citation_count":174,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"24144296","id":"PMC_24144296","title":"Nine loci for ocular axial length identified through genome-wide association studies, including shared loci with refractive error.","date":"2013","source":"American journal of human genetics","url":"https://pubmed.ncbi.nlm.nih.gov/24144296","citation_count":136,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"15116387","id":"PMC_15116387","title":"Expression pattern of lacZ reporter gene representing connexin36 in transgenic mice.","date":"2004","source":"The Journal of comparative neurology","url":"https://pubmed.ncbi.nlm.nih.gov/15116387","citation_count":104,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"19095792","id":"PMC_19095792","title":"The neuronal connexin36 interacts with and is phosphorylated by CaMKII in a way similar to CaMKII interaction with glutamate receptors.","date":"2008","source":"Proceedings of the National Academy of Sciences of the United States of America","url":"https://pubmed.ncbi.nlm.nih.gov/19095792","citation_count":99,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"32694731","id":"PMC_32694731","title":"Systematic mapping of genetic interactions for de novo fatty acid synthesis identifies C12orf49 as a regulator of lipid metabolism.","date":"2020","source":"Nature metabolism","url":"https://pubmed.ncbi.nlm.nih.gov/32694731","citation_count":92,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"16211561","id":"PMC_16211561","title":"Human and mouse microglia express connexin36, and functional gap junctions are formed between rodent microglia and neurons.","date":"2005","source":"Journal of neuroscience research","url":"https://pubmed.ncbi.nlm.nih.gov/16211561","citation_count":85,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"15235036","id":"PMC_15235036","title":"Association of the connexin36 gene with juvenile myoclonic epilepsy.","date":"2004","source":"Journal of medical genetics","url":"https://pubmed.ncbi.nlm.nih.gov/15235036","citation_count":60,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"14565956","id":"PMC_14565956","title":"Critical role of the transcriptional repressor neuron-restrictive silencer factor in the specific control of connexin36 in insulin-producing cell lines.","date":"2003","source":"The Journal of biological chemistry","url":"https://pubmed.ncbi.nlm.nih.gov/14565956","citation_count":59,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"17828386","id":"PMC_17828386","title":"Beta cells preferentially exchange cationic molecules via connexin 36 gap junction channels.","date":"2007","source":"Diabetologia","url":"https://pubmed.ncbi.nlm.nih.gov/17828386","citation_count":53,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"23756480","id":"PMC_23756480","title":"The quest for juvenile myoclonic epilepsy genes.","date":"2013","source":"Epilepsy & behavior : E&B","url":"https://pubmed.ncbi.nlm.nih.gov/23756480","citation_count":48,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"12522556","id":"PMC_12522556","title":"Mutations in the calcium-binding motifs of CDH23 and the 35delG mutation in GJB2 cause hearing loss in one family.","date":"2002","source":"Human genetics","url":"https://pubmed.ncbi.nlm.nih.gov/12522556","citation_count":40,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"16876983","id":"PMC_16876983","title":"Confirmatory evidence for an association of the connexin-36 gene with juvenile myoclonic epilepsy.","date":"2006","source":"Epilepsy research","url":"https://pubmed.ncbi.nlm.nih.gov/16876983","citation_count":37,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"19418635","id":"PMC_19418635","title":"Direct association of connexin36 with zonula occludens-2 and zonula occludens-3.","date":"2009","source":"Neurochemistry international","url":"https://pubmed.ncbi.nlm.nih.gov/19418635","citation_count":36,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"20979653","id":"PMC_20979653","title":"Expression of connexin genes in the human retina.","date":"2010","source":"BMC ophthalmology","url":"https://pubmed.ncbi.nlm.nih.gov/20979653","citation_count":36,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"25233373","id":"PMC_25233373","title":"Genome-wide meta-analysis of myopia and hyperopia provides evidence for replication of 11 loci.","date":"2014","source":"PloS one","url":"https://pubmed.ncbi.nlm.nih.gov/25233373","citation_count":36,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"36115835","id":"PMC_36115835","title":"Quantitative fragmentomics allow affinity mapping of interactomes.","date":"2022","source":"Nature communications","url":"https://pubmed.ncbi.nlm.nih.gov/36115835","citation_count":33,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"36906653","id":"PMC_36906653","title":"Cryo-EM structures of human Cx36/GJD2 neuronal gap junction channel.","date":"2023","source":"Nature communications","url":"https://pubmed.ncbi.nlm.nih.gov/36906653","citation_count":31,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"25138779","id":"PMC_25138779","title":"Genome-wide association study of maternal and inherited effects on left-sided cardiac malformations.","date":"2014","source":"Human molecular genetics","url":"https://pubmed.ncbi.nlm.nih.gov/25138779","citation_count":25,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"26173976","id":"PMC_26173976","title":"Characterization of connexin36 gap junctions in the human outer retina.","date":"2015","source":"Brain structure & function","url":"https://pubmed.ncbi.nlm.nih.gov/26173976","citation_count":25,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"22752717","id":"PMC_22752717","title":"Neurons and β-cells of the pancreas express connexin36, forming gap junction channels that exhibit strong cationic selectivity.","date":"2012","source":"The Journal of membrane biology","url":"https://pubmed.ncbi.nlm.nih.gov/22752717","citation_count":20,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"36724073","id":"PMC_36724073","title":"Sequential genome-wide CRISPR-Cas9 screens identify genes regulating cell-surface expression of tetraspanins.","date":"2023","source":"Cell reports","url":"https://pubmed.ncbi.nlm.nih.gov/36724073","citation_count":20,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"30295974","id":"PMC_30295974","title":"E3 ubiquitin ligases LNX1 and LNX2 localize at neuronal gap junctions formed by connexin36 in rodent brain and molecularly interact with connexin36.","date":"2018","source":"The European journal of neuroscience","url":"https://pubmed.ncbi.nlm.nih.gov/30295974","citation_count":17,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"26485405","id":"PMC_26485405","title":"Association between Ocular Axial Length-Related Genes and High Myopia in a Han Chinese Population.","date":"2015","source":"Ophthalmologica. Journal international d'ophtalmologie. International journal of ophthalmology. Zeitschrift fur Augenheilkunde","url":"https://pubmed.ncbi.nlm.nih.gov/26485405","citation_count":14,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"29246791","id":"PMC_29246791","title":"A potential role for neuronal connexin 36 in the pathogenesis of amyotrophic lateral sclerosis.","date":"2017","source":"Neuroscience letters","url":"https://pubmed.ncbi.nlm.nih.gov/29246791","citation_count":11,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"29793445","id":"PMC_29793445","title":"Heritability of myopia and its relation with GDJ2 and RASGRF1 genes in Lithuania.","date":"2018","source":"BMC ophthalmology","url":"https://pubmed.ncbi.nlm.nih.gov/29793445","citation_count":10,"is_preprint":false,"source_track":"gene2pubmed"},{"pmid":"27916369","id":"PMC_27916369","title":"Expression of connexin 26 and connexin 43 is reduced in Hirschsprung's disease.","date":"2016","source":"The Journal of surgical research","url":"https://pubmed.ncbi.nlm.nih.gov/27916369","citation_count":8,"is_preprint":false,"source_track":"gene2pubmed"}],"cost":{"stage1":{"model":"claude-sonnet-4-6","input_tokens":34974,"output_tokens":6326,"usd":0.099906},"stage2":{"model":"claude-opus-4-6","input_tokens":9971,"output_tokens":3515,"usd":0.206595},"total_usd":0.728466,"stage1_batch_id":"msgbatch_01DzrsgVgYngY1hvM3DU1yF2","stage2_batch_id":"msgbatch_01KA9b7iRzKFdhoBUT7vCvSe","note":"batch pricing = 50% of standard","round2_stage1":{"model":"claude-sonnet-4-6","input_tokens":48985,"output_tokens":10721,"usd":0.153885},"round2_rules_fired":"R2","round2_stage2":{"model":"claude-opus-4-6","input_tokens":14634,"output_tokens":4222,"usd":0.26808}},"stage1_raw":"```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 1998,\n      \"finding\": \"Cx36 (GJD2) was identified as a new gap junction protein of 321 amino acids predominantly expressed in mammalian neurons, particularly in the inferior olive, olfactory bulb, CA3/CA4 hippocampal subfields, and retina, establishing it as the first connexin expressed predominantly in mammalian neurons.\",\n      \"method\": \"Degenerate RT-PCR cloning, genomic library screening, in situ hybridization, neurotoxin lesion experiments\",\n      \"journal\": \"The European journal of neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1/2 — original cloning with multiple orthogonal methods, foundational paper, 377 citations\",\n      \"pmids\": [\"9753189\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 1999,\n      \"finding\": \"The human CX36 gene (GJD2) was cloned, shown to have 98% protein identity with mouse/rat Cx36, contains a single intron 71 bp after the translation initiation site, and was chromosomally localized to band 15q14.\",\n      \"method\": \"Gene cloning, sequencing, fluorescence in situ hybridization (FISH), radioactive in situ hybridization\",\n      \"journal\": \"Journal of neuroscience research\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — direct cloning and chromosomal localization with multiple methods\",\n      \"pmids\": [\"10462698\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2000,\n      \"finding\": \"Cx36 is selectively expressed by insulin-producing beta-cells within pancreatic islets, not by other endocrine cell types, establishing Cx36 as the connexin mediating beta-cell coupling.\",\n      \"method\": \"In situ hybridization, immunolabeling, FACS-purified beta vs. non-beta cell preparations\",\n      \"journal\": \"Diabetes\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — multiple orthogonal methods, replicated by subsequent studies\",\n      \"pmids\": [\"10905480\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2001,\n      \"finding\": \"Using freeze-fracture replica immunogold labeling (FRIL), Cx36 was found exclusively in neuronal gap junctions in adult rat CNS, whereas Cx26, Cx30, and Cx43 were restricted to astrocytes and Cx32 to oligodendrocytes, defining separate gap junction communication pathways.\",\n      \"method\": \"Freeze-fracture replica immunogold labeling (FRIL), confocal immunocytochemistry\",\n      \"journal\": \"Cell communication & adhesion\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — ultrastructural co-localization with 100-fold higher resolution than confocal, >4000 labeled junctions examined\",\n      \"pmids\": [\"12064610\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2001,\n      \"finding\": \"Loss of Cx36 in beta-cells does not affect development or differentiation but alters insulin secretion, and Cx36 expression changes are paralleled by changes in beta-cell function in insulinoma-bearing rats, establishing a functional role for Cx36 in insulin secretion regulation.\",\n      \"method\": \"Cx36 knockout mice, antisense transfection of MIN6 cells, insulin secretion assays\",\n      \"journal\": \"Cell communication & adhesion\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — KO mouse phenotype plus cell-line loss-of-function with defined secretory readout, replicated across studies\",\n      \"pmids\": [\"12064624\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2002,\n      \"finding\": \"The N-terminal domain of Cx36 is required for normal subcellular localization; mutagenesis of the N-terminus disrupts protein trafficking, while mutagenesis of putative phosphorylation motifs does not alter localization, excluding phosphorylation as a major step in Cx36 protein transport.\",\n      \"method\": \"Cx36-EGFP expression in neuroblastoma cells and primary hippocampal neurons, site-directed mutagenesis, electrophysiology\",\n      \"journal\": \"Journal of neuroscience research\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — direct mutagenesis with localization and functional readout, single lab\",\n      \"pmids\": [\"12210839\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2003,\n      \"finding\": \"Cx36 modulates insulin secretion by controlling cytosolic Ca2+ handling in populations of beta-cells; loss of Cx36 expression alters Ca2+ dynamics in insulin-producing cells, identifying a molecular link between Cx36 and the stimulus-secretion coupling pathway.\",\n      \"method\": \"Antisense Cx36 transfection, cytosolic Ca2+ monitoring, insulin secretion assays\",\n      \"journal\": \"Cell communication & adhesion\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — loss-of-function with defined Ca2+ and secretory readout, single lab\",\n      \"pmids\": [\"14681053\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2004,\n      \"finding\": \"Cx36 (but not E-cadherin) is specifically required for normal insulin secretion steps in MIN6 beta-cells; restoring Cx36 expression at adequate but not excessive levels rescues insulin secretion in Cx36-knockdown cells, while E-cadherin restoration fails to do so.\",\n      \"method\": \"Lentiviral transduction to restore Cx36 or E-cadherin expression in antisense Cx36-transfected MIN6 cells, insulin secretion assays\",\n      \"journal\": \"Experimental cell research\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — rescue experiment with specific isoform controls establishing causality, moderate evidence\",\n      \"pmids\": [\"15023528\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2004,\n      \"finding\": \"Mefloquine potently and selectively blocks Cx36 gap junction channels (IC50 ~300 nM) and Cx50 channels (IC50 ~1.1 µM) in transfected N2A cells, while other connexins (Cx43, Cx32, Cx26) are only affected at 10–100-fold higher concentrations; mefloquine blocks interneuronal coupling in neocortical slices at 25 µM.\",\n      \"method\": \"Transfection of N2A neuroblastoma cells, electrophysiology (dual patch-clamp), acute brain slice recordings\",\n      \"journal\": \"Proceedings of the National Academy of Sciences of the United States of America\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — in vitro channel assay in transfected cells plus slice validation, 271 citations\",\n      \"pmids\": [\"15297615\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2007,\n      \"finding\": \"Cx36-mediated electrical coupling between beta-cells hyperpolarizes active cells via coupled inactive neighbors; loss of Cx36 causes continuous electrical activity when KATP channels are blocked, and absence of electrical coupling modifies electrophysiological properties including increased Ca2+ channel density, affecting stimulation threshold and kinetics of insulin release.\",\n      \"method\": \"Electrophysiological recordings in pancreatic slices from Cx36-/-, Cx36+/-, and Cx36+/+ mice\",\n      \"journal\": \"Diabetes\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — genetic KO with direct electrophysiological readout, multiple genotypes compared\",\n      \"pmids\": [\"17395748\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2008,\n      \"finding\": \"Human beta-cells express Cx36 protein concentrated in lipid raft domains of the cell membrane; Cx36 channels allow preferential exchange of cationic molecules between human beta-cells, and Cx36 mRNA levels correlate with insulin gene expression.\",\n      \"method\": \"Immunolabeling, lipid raft fractionation, dye transfer assays, mRNA correlation analysis\",\n      \"journal\": \"Human molecular genetics\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — multiple orthogonal methods establishing subcellular localization and channel permeability properties\",\n      \"pmids\": [\"19000992\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2012,\n      \"finding\": \"Beta2/NeuroD1 transcription factor binds E-box elements in the Gjd2 (Cx36) promoter and, together with the E47 cofactor, transactivates the Gjd2 promoter, establishing Cx36 as a transcriptional target of Beta2/NeuroD1 in beta-cell development.\",\n      \"method\": \"Promoter-reporter assays, chromatin immunoprecipitation (ChIP), transcription factor binding analysis\",\n      \"journal\": \"The Journal of membrane biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — direct binding to promoter E-boxes demonstrated with ChIP plus functional transactivation assay\",\n      \"pmids\": [\"22729650\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2014,\n      \"finding\": \"Genetic knockout of Cx36 in mice expressing ATP-insensitive KATP channels rescues glucose-stimulated Ca2+ activity and insulin secretion, preventing hyperglycemia; this demonstrates that Cx36 gap junctions suppress beta-cell excitability and that reduced coupling can compensate for overactive KATP channels.\",\n      \"method\": \"Genetic double knockout mouse model, Ca2+ imaging, insulin secretion assays, glucose tolerance tests\",\n      \"journal\": \"Diabetes\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — genetic epistasis with defined mechanistic rescue, multiple physiological readouts\",\n      \"pmids\": [\"24458355\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2015,\n      \"finding\": \"Cx36 phosphorylation at Ser293 is significantly increased in AII amacrine cells of two retinal degeneration mouse models (rd10 and rd1), suggesting that elevated phosphorylation underlies increased gap junction coupling observed in degenerating retinas.\",\n      \"method\": \"Immunofluorescence with phospho-specific Ser293-P antibody in rd10 and rd1 mouse retinas\",\n      \"journal\": \"Frontiers in cellular neuroscience\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 — phosphorylation state detected by immunofluorescence, single lab, no direct functional mutagenesis\",\n      \"pmids\": [\"26483638\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"A synonymous SNP (rs3743123) in human GJD2 alters gap junction plaque formation and cell coupling in HeLa cells, and transgenic mice expressing this variant show postnatal reduction of islet Cx36 levels and beta-cell survival with hyperglycemia, demonstrating that a silent coding polymorphism can functionally impair Cx36-dependent beta-cell coupling.\",\n      \"method\": \"HeLa cell transfection, gap junction plaque imaging, coupling assays, transgenic mouse lines with insulin-promoter-driven expression\",\n      \"journal\": \"PloS one\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — cell-based and in vivo transgenic validation with multiple phenotypic readouts\",\n      \"pmids\": [\"26959991\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"Ca2+-loaded calmodulin (CaM) binds the C-terminus of Cx36 in a compact conformation with major hydrophobic contributions from W277 and V284; CaM binding is Ca2+-dependent, activates Cx36 channels (unlike other connexins), and competes with CaMKII for an overlapping binding site, establishing Cx36 as a hub for Ca2+/CaM-mediated channel regulation.\",\n      \"method\": \"NMR solution structure of CaM-Cx36 peptide complex, co-immunoprecipitation, electrophysiology, Ca2+-dependence assays\",\n      \"journal\": \"Frontiers in molecular neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — NMR structure plus functional validation by electrophysiology, identifies specific binding residues\",\n      \"pmids\": [\"27917108\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2017,\n      \"finding\": \"CaMKII-β and CaMKII-δ isoforms co-localize specifically with Cx36-containing gap junctions in the mouse retina; CaMKII-β colocalizes with Cx36 in both inner and outer retina, while CaMKII-δ only in the inner retina (AII amacrine cells), suggesting isoform-specific regulation of electrical synapses.\",\n      \"method\": \"Confocal microscopy, immunofluorescence co-localization in mouse retina with isoform-specific antibodies\",\n      \"journal\": \"Frontiers in molecular neuroscience\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 — immunofluorescence co-localization, single lab, no direct functional coupling measurement\",\n      \"pmids\": [\"29311815\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2018,\n      \"finding\": \"Cx36 gap junction conductance is stimulated by short-chain n-alcohols and isoflurane via a pocket near three unique cysteine residues (C264 in TM4, C92 and C87 in TM2); substitution of these cysteines reversed the stimulatory effect to inhibitory, suggesting that re-shuffling of inter-subunit disulphide bonds mediates stimulation.\",\n      \"method\": \"Structural modeling, cysteine mutagenesis, dual whole-cell patch-clamp in HeLa and N2A cells\",\n      \"journal\": \"Bioscience reports\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — mutagenesis combined with in vitro electrophysiology identifies specific structural mechanism\",\n      \"pmids\": [\"29298877\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2020,\n      \"finding\": \"Local Ca2+ signaling at Cx36 gap junctions can be measured with a Cx36-GCaMP fusion biosensor; glutamate receptor activation (including endogenous NMDA/kainate receptors in HeLa cells) increases local Ca2+ at Cx36 junctions and potentiates Cx36 coupling via CaMKII activity, demonstrating glutamate-receptor-driven potentiation of electrical synapses.\",\n      \"method\": \"Cx36-GCaMP fusion construct in transfected cells, Ca2+ imaging, tracer coupling assays, pharmacological CaMKII inhibition\",\n      \"journal\": \"eNeuro\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1/2 — engineered biosensor with functional coupling readout and pharmacological validation\",\n      \"pmids\": [\"32179580\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2021,\n      \"finding\": \"Loss of gjd2a (Cx35.5) in zebrafish leads to hyperopia and electrophysiological changes in the retina, while loss of gjd2b (Cx35.1) causes nuclear cataract that triggers axial elongation, demonstrating distinct roles of GJD2 orthologs in ocular biometry and refractive error development.\",\n      \"method\": \"Zebrafish knockout/knockdown, immunohistochemistry, scRNA sequencing, retinal electrophysiology, biometry\",\n      \"journal\": \"Communications biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — loss-of-function in zebrafish with multiple orthogonal readouts, GJD2 ortholog\",\n      \"pmids\": [\"34083742\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"Cx36 trafficking from the ER to the cis-Golgi requires the C-terminal valine as an ER export signal that recruits COPII cargo receptors Sec24A/B/C at ER exit sites; the PDZ binding motif 'SAYV' mediates interaction with Golgi stacking protein Grasp55, which stabilizes Cx36 in the Golgi, while Sec24 subunits carry Cx36 out of the ER.\",\n      \"method\": \"siRNA knockdown, BioID proximity labeling screens, overexpression experiments, confocal microscopy in HEK293T cells\",\n      \"journal\": \"Cellular and molecular life sciences : CMLS\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1/2 — BioID interactome plus siRNA functional validation identifies specific trafficking machinery\",\n      \"pmids\": [\"39395036\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"Cryo-EM structures of human Cx36 gap junction channel in complex with mefloquine, arachidonic acid, and 1-hexanol reveal that these inhibitors competitively bind to the N-terminal helix (NTH) binding pocket, inducing a conformational shift from the pore-lining NTH (PLN) state to a flexible NTH (FN) state, leading to obstruction of the channel pore by lipid bilayer densities.\",\n      \"method\": \"Cryo-electron microscopy structure determination of human Cx36 GJC with inhibitors\",\n      \"journal\": \"Nature communications\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — cryo-EM structure with multiple inhibitors revealing molecular gating mechanism\",\n      \"pmids\": [\"39455592\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"At least three electrostatic interaction (ESI) residue pairs per E2-E2 interface are required for functional Cx36 gap junction formation; these unique ESIs ensure Cx36 docking specificity to itself, preventing heterotypic gap junction formation with other brain connexins (Cx26, Cx30, Cx31.3, Cx32, Cx43, Cx45, Cx47).\",\n      \"method\": \"Missense variant design, computational modeling, dual patch-clamp electrophysiology in engineered HEK293 cell pairs\",\n      \"journal\": \"bioRxiv\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — mutagenesis combined with electrophysiology, informed by crystal structure ESI pairs\",\n      \"pmids\": [],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"ZO-1 interacts with Cx36 via its PDZ1 domain through the C-terminal PDZ binding motif 'SAYV' of Cx36; this interaction is inherently weak and sensitive to CaMKII-mediated phosphorylation of Cx36, suggesting ZO-1 unbinding is a necessary event for potentiation of electrical synapses; acidic residues adjacent to the PBM evolutionarily tune binding affinity.\",\n      \"method\": \"Gaussian accelerated molecular dynamics simulations, binding assays, site-directed mutagenesis\",\n      \"journal\": \"bioRxiv\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — MD simulation plus binding assays, preprint not yet peer-reviewed\",\n      \"pmids\": [],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"In vivo BioID proximity labeling in retinal AII amacrine cells identified >50 new Cx36-associated proteins including scaffold proteins, adhesion molecules, and cytoskeletal regulators; Sipa1l3 was identified as a new electrical synapse scaffold that interacts with ZO-1, ZO-2, and Cx36.\",\n      \"method\": \"In vivo BioID proximity labeling in mouse AII amacrine cells and transgenic zebrafish, co-IP binding assays, immunofluorescence\",\n      \"journal\": \"bioRxiv\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — in vivo proximity labeling with binding validation, preprint\",\n      \"pmids\": [],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"Cx36 hemichannels allow release of ATP but not glutamate or lactate, establishing size-independent selective permeability; the N-terminus charge and N-terminus–TM2 interactions are key structural determinants of this permselectivity.\",\n      \"method\": \"Co-expression of genetically encoded fluorescent sensors with Cx36, mutational analysis, patch-clamp\",\n      \"journal\": \"bioRxiv\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 1 — in vitro assay with mutagenesis, but preprint\",\n      \"pmids\": [],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"D1 receptor signaling regulates AII amacrine cell membrane potential and glycine release to OFF cone bipolar cells by modulating Cx36 gap junction coupling; D1R antagonists depolarize AII amacrine cells and enhance glycinergic transmission in wild-type but not Cx36-/- mice.\",\n      \"method\": \"Whole-cell patch clamp in retinal slices from wild-type and Cx36-/- mice, pharmacological D1R manipulation\",\n      \"journal\": \"bioRxiv\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — KO vs wild-type comparison with pharmacological validation, preprint\",\n      \"pmids\": [],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"Cx36 KO mice display reduced fine odor discrimination for structurally similar odor pairs and reduced late excitation of mitral cells in olfactory bulb slices, demonstrating that Cx36-mediated electrical synapses contribute to fine odor processing and mitral cell excitation.\",\n      \"method\": \"Go/no-go behavioral odor discrimination task in Cx36 KO mice, electrophysiological recordings in OB slices\",\n      \"journal\": \"bioRxiv\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — KO behavioral and electrophysiological phenotype, preprint\",\n      \"pmids\": [],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"Delta cell coordination with beta cells is mediated by a combination of paracrine signaling (via vesicle release) and low-density Cx36 gap junctions; selective Cx36 knockout in delta cells reduced their coupling to beta cells, and blockade of vesicle release removed coupling between beta and most delta cells.\",\n      \"method\": \"GCaMP6s Ca2+ imaging of hundreds of beta and delta cells, selective delta cell Cx36 KO, ML-141 vesicle release blockade\",\n      \"journal\": \"Proceedings of the National Academy of Sciences of the United States of America\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — cell-type-specific KO with live Ca2+ imaging and pharmacological dissection, peer-reviewed\",\n      \"pmids\": [\"40956879\"],\n      \"is_preprint\": false\n    }\n  ],\n  \"current_model\": \"GJD2/Cx36 is the principal neuronal and pancreatic beta-cell connexin that assembles into gap junction channels forming electrical synapses; it is trafficked from the ER via COPII/Sec24-dependent export and stabilized in the Golgi by Grasp55, localized to the plasma membrane where its channel conductance is gated by N-terminal helix conformational changes (targeted by inhibitors such as mefloquine, arachidonic acid, and alcohols via unique cysteine residues in TM2/TM4), regulated by CaM/CaMKII-dependent phosphorylation (particularly at Ser293) and anchored by ZO-1 at electrical synapses, with electrostatic interactions between E2 loops ensuring homospecific docking; in neurons it synchronizes network activity and fine odor discrimination, and in beta-cells it coordinates Ca2+ oscillations and insulin secretion through electrical coupling that also integrates paracrine signals from delta cells.\"\n}\n```\n\n--- ROUND 2 (R2/R3, augmented corpus) ---\n\nI'll classify each paper first, then extract discoveries from KEEP papers.\n\n**Key exclusions:**\n- Paper [21] (GJA9/connexin59 in dogs) — alias collision, different gene\n- Paper [59] (lncRNA GJA9-MYCBP) — alt-locus product / different gene\n- Paper [15] (zebrafish cx36.7/Ecx cardiac connexin) — symbol collision; this is a distinct zebrafish connexin with different function (cardiac morphogenesis via Nkx2.5), not an ortholog of mammalian GJD2/Cx36\n- Paper [63] (zebrafish cx36.7 promoter) — same as above, cardiac zebrafish connexin\n- Gene2pubmed papers [9] (C12orf49/fatty acid synthesis), [25] (tetraspanin CRISPR screens) — clearly off-target\n- Papers that are pure GWAS/association/expression without mechanism: [31], [38], [54], [60] (LOT kidney mutation report only)\n\nAll other papers about Cx36/GJD2 (neuronal gap junction protein) are KEEP.\n\n```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 1998,\n      \"finding\": \"Cx36 (encoded by GJD2) was cloned as a new connexin family member of 321 amino acids, and in situ hybridization showed it is expressed predominantly in mammalian neurons (inferior olive, olfactory bulb, hippocampus, brainstem nuclei, retinal ganglion cell and inner nuclear layers), making it the first connexin shown to be expressed preferentially in neurons rather than glia.\",\n      \"method\": \"Degenerate RT-PCR cloning, genomic library isolation, in situ hybridization, neurotoxin lesion experiments\",\n      \"journal\": \"The European journal of neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — original cloning with multiple orthogonal methods and neurotoxin validation, highly cited foundational paper\",\n      \"pmids\": [\"9753189\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 1999,\n      \"finding\": \"The human GJD2 gene was cloned and localized to chromosome 15q14; its coding sequence is 98% identical at the protein level to mouse/rat Cx36, and the gene structure (single intron 71 bp after the translation initiation site) is conserved across species, confirming it is the human ortholog of rodent Cx36.\",\n      \"method\": \"Gene cloning, sequencing, fluorescence in situ hybridization (FISH), radioactive in situ hybridization\",\n      \"journal\": \"Journal of neuroscience research\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — direct cloning and chromosomal localization with multiple methods\",\n      \"pmids\": [\"10462698\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2000,\n      \"finding\": \"Cx36 is selectively expressed by insulin-producing beta-cells in the central region of pancreatic islets, with little or no expression in other endocrine cell types (alpha, delta cells), establishing beta-cell-specific gap junctional coupling via Cx36.\",\n      \"method\": \"In situ hybridization, immunolabeling with affinity-purified antibodies, FACS-purified beta- vs. non-beta-cell fractions\",\n      \"journal\": \"Diabetes\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — multiple orthogonal methods, replicated across subsequent studies\",\n      \"pmids\": [\"10905480\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2001,\n      \"finding\": \"Freeze-fracture replica immunogold labeling (FRIL) demonstrated that Cx36 is exclusively localized to neuronal gap junctions in adult rat CNS, whereas Cx26, Cx30, and Cx43 are restricted to astrocytes and Cx32 to oligodendrocytes, establishing separate and non-overlapping connexin-defined pathways for neuronal versus glial gap junctional communication.\",\n      \"method\": \"Freeze-fracture replica immunogold labeling (FRIL), confocal immunocytochemistry\",\n      \"journal\": \"Cell communication & adhesion\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — ultrastructural immunogold co-localization in >4000 labeled gap junctions in >370 replicas\",\n      \"pmids\": [\"12064610\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2001,\n      \"finding\": \"Global ischemia induced a selective increase in Cx36 protein (without corresponding mRNA increase) in parvalbumin-positive inhibitory interneurons of CA1 hippocampus before neuronal death, suggesting post-translational regulation of Cx36 and a role for Cx36 gap junctions in the survival of GABAergic interneurons.\",\n      \"method\": \"Western blot, Northern blot, in situ hybridization, double immunofluorescence, Cx32 knockout mouse model\",\n      \"journal\": \"The Journal of neuroscience\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — multiple methods in one lab, but mechanistic link to survival is inferred from KO phenotype\",\n      \"pmids\": [\"11567043\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2002,\n      \"finding\": \"N-terminal domain mutagenesis of Cx36 identified a structural element necessary for normal subcellular localization; site-directed mutagenesis of putative phosphorylation motifs did not alter localization, indicating that phosphorylation/dephosphorylation is not the major regulatory step in Cx36 protein transport. Cx36-EGFP formed functional gap junction plaques with electrical properties indistinguishable from native Cx36.\",\n      \"method\": \"Cx36-EGFP transfection in neuroblastoma cell lines and primary hippocampal neurons, mutagenesis, electrophysiology\",\n      \"journal\": \"Journal of neuroscience research\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — mutagenesis combined with functional electrophysiology and imaging\",\n      \"pmids\": [\"12210839\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2003,\n      \"finding\": \"The neuron-restrictive silencer factor (NRSF/REST) acts as a transcriptional repressor of the GJD2 (Cx36) gene via a conserved neuron-restrictive silencer element (NRSE) in the promoter; ectopic NRSF expression in beta-cell lines markedly reduced Cx36 mRNA and protein, and mutation of the NRSE relieved repression, explaining why Cx36 is expressed specifically in neurons and beta-cells (which lack NRSF).\",\n      \"method\": \"Luciferase reporter assays, viral gene transfer of NRSF into beta-cell lines, NRSE mutagenesis, trichostatin A (HDAC inhibitor) treatment\",\n      \"journal\": \"The Journal of biological chemistry\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — promoter dissection with mutagenesis, gain-of-function, and pharmacological validation across multiple cell lines\",\n      \"pmids\": [\"14565956\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2003,\n      \"finding\": \"Cx36-dependent gap junction coupling controls cytosolic Ca2+ handling in populations of insulin-producing cells; loss of Cx36 expression altered Ca2+ transients, identifying a molecular link between Cx36 and the stimulus-secretion pathway for insulin secretion.\",\n      \"method\": \"Antisense Cx36 transfection in MIN6 cells, Ca2+ imaging, insulin secretion assays\",\n      \"journal\": \"Cell communication & adhesion\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — functional Ca2+ imaging linked to Cx36 loss-of-function with secretion readout\",\n      \"pmids\": [\"14681053\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2004,\n      \"finding\": \"Mefloquine potently and selectively blocks Cx36 gap junction channels (IC50 ~300 nM) and Cx50 channels (IC50 ~1.1 µM) while only affecting Cx43, Cx32, and Cx26 at 10–100-fold higher concentrations; mefloquine (25 µM) blocked gap junctional coupling between neocortical interneurons in brain slices with minimal nonspecific effects on synaptic potentials or intrinsic properties.\",\n      \"method\": \"Whole-cell patch clamp in transfected N2A cells, acute neocortical brain slices, pharmacological dose-response\",\n      \"journal\": \"Proceedings of the National Academy of Sciences of the United States of America\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — in vitro electrophysiology with dose-response and in situ slice validation; replicated in subsequent studies\",\n      \"pmids\": [\"15297615\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2004,\n      \"finding\": \"Cx36, but not E-cadherin, is specifically required for normal insulin secretion in MIN6 beta-cells; lentivirus-mediated restoration of Cx36 (but not E-cadherin) rescued insulin secretory defects in Cx36 antisense-transfected clones, demonstrating that Cx36 gap junction function specifically controls beta-cell secretion.\",\n      \"method\": \"Stable antisense transfection, lentiviral transduction, insulin secretion assays\",\n      \"journal\": \"Experimental cell research\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — rescue experiment with specific connexin vs. control protein, multiple secretagogues tested\",\n      \"pmids\": [\"15023528\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2004,\n      \"finding\": \"A lacZ reporter driven by the Cx36 locus revealed expression in GABAergic neurons of cerebellar nuclei, non-GABAergic neurons of the inferior olive, mitral cells of the olfactory bulb, parvalbumin-positive cells of cerebral cortex, retinal layers, insulin-producing beta-cells of the pancreas, and the adrenal medulla, establishing the full tissue distribution of Cx36 in adult mice.\",\n      \"method\": \"Beta-galactosidase reporter gene expression in Cx36(+/del[LacZ]) transgenic mice, histochemistry\",\n      \"journal\": \"The Journal of comparative neurology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — genetic reporter strategy provides definitive cell-type attribution; highly cited\",\n      \"pmids\": [\"15116387\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2005,\n      \"finding\": \"Human and mouse microglia express Cx36 mRNA and protein, and Cx36 gap junctions form between microglia and neurons in co-culture, providing electrophysiological coupling below 30 pS consistent with Cx36 channel properties; this establishes a direct physical channel of communication between microglia and neurons via Cx36.\",\n      \"method\": \"RT-PCR, Western blot, immunofluorescence, electrophysiology (coupling measurements), Lucifer yellow dye coupling\",\n      \"journal\": \"Journal of neuroscience research\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — multiple methods from single lab including electrophysiology\",\n      \"pmids\": [\"16211561\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2006,\n      \"finding\": \"FRIL ultrastructural analysis confirmed that Cx36 (and less abundantly Cx45) is present at neuronal gap junctions specifically at 'mixed' glutamatergic/electrical synapses between mitral/tufted cell dendrites in the olfactory bulb; Cx36 was absent from olfactory receptor neuron axons, and genomic analysis revealed multiple miRNA binding sites in the 3'-UTR of Cx36 consistent with post-transcriptional regulation.\",\n      \"method\": \"Freeze-fracture replica immunogold labeling (FRIL), confocal immunofluorescence, genomic miRNA binding site analysis\",\n      \"journal\": \"Journal of neurocytology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — ultrastructural FRIL co-localization at defined synapse types\",\n      \"pmids\": [\"16841170\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2007,\n      \"finding\": \"Cx36 gap junctional coupling electrically hyperpolarizes beta-cells by propagating the membrane potential of adjacent inactive cells; in Cx36 knockout mice with blocked KATP channels, beta-cells rapidly depolarized and showed continuous electrical activity rather than the hyperpolarized state seen in coupled cells. Loss of Cx36 also increased Ca2+ channel density and altered the kinetics of insulin secretion in response to glucose changes.\",\n      \"method\": \"Electrophysiology in pancreatic slices from Cx36-/-, Cx36+/-, and Cx36+/+ mice; Ca2+ channel recordings; comparison with NMRI mice\",\n      \"journal\": \"Diabetes\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — direct electrophysiology in genetic knockout model with multiple functional readouts\",\n      \"pmids\": [\"17395748\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2007,\n      \"finding\": \"Cx36 channels in beta-cells preferentially permit exchange of cationic molecules (cations and positively charged tracers) over anionic molecules, demonstrating strong cation selectivity; glibenclamide stimulation increased coupling for all permeant molecules, while coupling extent varied depending on molecular charge.\",\n      \"method\": \"Microinjection of tracers differing in size and charge into islet cells of Cx36 knockout, transgenic, and wild-type mice; MIN6 cell coupling assays\",\n      \"journal\": \"Diabetologia\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — systematic permeability study using multiple tracers across genetic models\",\n      \"pmids\": [\"17828386\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2008,\n      \"finding\": \"CaMKII directly binds to two juxtamembrane cytoplasmic domains of Cx36 and phosphorylates Cx36 in vitro; both binding domains resemble the pseudosubstrate and pseudotarget sites of CaMKII and show phosphorylation-dependent interaction and autonomous CaMKII activation, analogous to CaMKII interaction with the NR2B subunit of NMDA receptors. CaMKII and Cx36 were significantly co-localized in the inferior olive.\",\n      \"method\": \"35S-labeled CaMKII binding assay to Cx36 cytoplasmic domains, in vitro phosphorylation assay, colocalization by immunofluorescence in inferior olive\",\n      \"journal\": \"Proceedings of the National Academy of Sciences of the United States of America\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — in vitro binding and phosphorylation with domain mapping; mechanistic analogy to NMDA receptor validated by multiple methods\",\n      \"pmids\": [\"19095792\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2008,\n      \"finding\": \"Human beta-cells express Cx36 protein at the membrane in lipid raft domains where it forms functional gap junctions that preferentially exchange cationic molecules; Cx36 mRNA levels correlate with insulin gene expression in both control and type 2 diabetic islets, establishing Cx36 as the native coupling protein of human beta-cells that contributes to gene expression regulation.\",\n      \"method\": \"Immunostaining, lipid raft fractionation, dye coupling assays, correlation analysis of Cx36 and insulin mRNA in human islets from T2D and control donors\",\n      \"journal\": \"Human molecular genetics\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — lipid raft localization plus functional coupling in human tissue; multiple methods\",\n      \"pmids\": [\"19000992\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2009,\n      \"finding\": \"Cx36 directly interacts with the PDZ1 domain of ZO-2 and ZO-3 via its C-terminal SAYV PDZ-binding motif; co-IP from Cx36-transfected HeLa cells and betaTC-3 cells confirmed co-immunoprecipitation, and in vitro pull-down with truncated Cx36 lacking SAYV abolished binding, establishing ZO-2 and ZO-3 as scaffold partners at Cx36-containing gap junctions.\",\n      \"method\": \"Co-immunoprecipitation, in vitro PDZ domain pull-down assay, C-terminal peptide competition, truncation mutagenesis\",\n      \"journal\": \"Neurochemistry international\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — reciprocal co-IP plus in vitro pull-down with domain and motif mutagenesis\",\n      \"pmids\": [\"19418635\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2012,\n      \"finding\": \"The transcription factor Beta2/NeuroD1 binds to three E-boxes in a 2-kbp fragment of the GJD2 promoter and, together with its cofactor E47, transactivates the promoter; this establishes Cx36 as a direct transcriptional target of Beta2/NeuroD1 during prenatal beta-cell differentiation, and Cx36 protein is selectively expressed by beta-cells throughout prenatal pancreas development.\",\n      \"method\": \"Reporter gene assay with GJD2 promoter fragments, ChIP-like binding assays, lentiviral expression of Beta2/NeuroD1, mouse pancreas developmental staging by immunolabeling\",\n      \"journal\": \"The Journal of membrane biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1-2 — promoter dissection with E-box binding confirmation and gain-of-function transactivation assay\",\n      \"pmids\": [\"22729650\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2012,\n      \"finding\": \"Cx36 gap junction channels exhibit ~10-fold higher single-channel permeability (Pγ) for cationic dyes vs. anionic dyes of similar mass; Pγ for anionic dye Alexa-350 was 358-fold lower than Cx43, 138-fold lower than Cx40, confirming that Cx36 is one of the most cation-selective connexins, consistent with its role in K+-mediated electrical coupling.\",\n      \"method\": \"Dual whole-cell voltage clamp combined with dual-mode fluorescence imaging of single gap junction channels in HeLa cells stably expressing Cx36-EGFP\",\n      \"journal\": \"The Journal of membrane biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — single-channel permeability measurements with rigorous quantitative comparison across connexin isoforms\",\n      \"pmids\": [\"22752717\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2012,\n      \"finding\": \"Intercellular synchronization of Ca2+ oscillations in MIN6 cells is a reliable measure of Cx36-dependent coupling; cells with reduced Cx36 show decreased Ca2+ synchrony, glibenclamide promotes Cx36 coupling and increases synchrony, and quinine inhibits it. Drug screens using this assay identified compounds affecting Cx36 distribution and insulin content.\",\n      \"method\": \"Semi-automatic fluorimetric Ca2+ imaging in MIN6 cell populations, pharmacological modulation of Cx36\",\n      \"journal\": \"PloS one\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — functional coupling assay validated against genetic and pharmacological Cx36 perturbations\",\n      \"pmids\": [\"22848521\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2014,\n      \"finding\": \"Genetic knockout of Cx36 in mice expressing ATP-insensitive KATP channels (neonatal diabetes model) restored near-normal glucose-stimulated Ca2+ activity and insulin secretion, and prevented hyperglycemia; demonstrating that Cx36-mediated coupling propagates the hyperpolarizing effect of overactive KATP channels across beta-cells, and that reducing coupling can compensate for this defect.\",\n      \"method\": \"Double-transgenic mouse model (Cx36 KO × ATP-insensitive KATP), glucose homeostasis measurements, islet Ca2+ imaging, insulin secretion assays\",\n      \"journal\": \"Diabetes\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — genetic epistasis experiment (double KO/transgenic) with multiple functional readouts\",\n      \"pmids\": [\"24458355\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2015,\n      \"finding\": \"Phosphorylation of Cx36 at Ser293 is significantly elevated in AII amacrine cells of rd10 and rd1 retinal degeneration mouse models compared to wildtype, using a phospho-specific antibody; this elevated phosphorylation state is associated with increased gap junction coupling that underlies aberrant spontaneous hyperactivity in degenerating retina.\",\n      \"method\": \"Immunofluorescence with phospho-Ser293 specific antibody in rd10 and rd1 mouse retinas vs. wildtype\",\n      \"journal\": \"Frontiers in cellular neuroscience\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 — single lab, single method (phospho-specific antibody) with functional inference\",\n      \"pmids\": [\"26483638\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"Calmodulin (CaM) binds to the carboxy-terminus of Cx36 in a calcium-dependent manner at a site overlapping the CaMKII binding site; NMR solution structure reveals CaM binds Cx36 in its characteristic compact state with major hydrophobic contacts at W277 (anchor position 1) and V284 (position 8); Ca2+-loaded CaM activates Cx36 channels, unlike its effect on other connexins.\",\n      \"method\": \"NMR spectroscopy, Ca2+-dependent binding assays, electrophysiology of Cx36 channel activity after CaM application\",\n      \"journal\": \"Frontiers in molecular neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — NMR structure of Cx36-CaM complex with functional validation of channel activation\",\n      \"pmids\": [\"27917108\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"A synonymous SNP (rs3743123, S196S) in GJD2 alters gap junction plaque formation and cell coupling in HeLa cells transfected with the variant cDNA compared to wildtype; transgenic mice expressing the variant under insulin promoter showed postnatal reduction of islet Cx36 levels and beta-cell survival, causing hyperglycemia, demonstrating functional consequences of this silent polymorphism.\",\n      \"method\": \"Transfection of HeLa cells, transgenic mouse lines with insulin promoter-driven variant vs. WT Cx36, immunostaining, glucose measurements\",\n      \"journal\": \"PloS one\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — cell and in vivo functional studies but single lab\",\n      \"pmids\": [\"26959991\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2017,\n      \"finding\": \"CaMKII-β and CaMKII-δ co-localize specifically with Cx36-containing gap junctions in the mouse retina; in the outer retina only CaMKII-β co-localizes with Cx36, while in the inner retina both CaMKII-β and -δ co-localize with Cx36, suggesting isoform-specific regulation of electrical synapses in different retinal layers.\",\n      \"method\": \"Confocal microscopy with double-labeling for Cx36 and four CaMKII isoforms in mouse retina\",\n      \"journal\": \"Frontiers in molecular neuroscience\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 3 — single lab, co-localization only, no direct functional interaction test\",\n      \"pmids\": [\"29311815\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2018,\n      \"finding\": \"E3 ubiquitin ligases LNX1 and LNX2 co-localize with Cx36-containing gap junctions in adult mouse brain, co-immunoprecipitate with Cx36, and LNX PDZ2 domain pulls down Cx36; cotransfection of E3 ligase-competent LNX isoforms with Cx36 caused loss of Cx36 gap junction plaques between cells, whereas ligase-inactive isoforms did not, establishing LNX-mediated ubiquitination as a mechanism for Cx36 internalization.\",\n      \"method\": \"Immunofluorescence colocalization, co-immunoprecipitation, PDZ pull-down, cotransfection of wild-type vs. ligase-dead LNX isoforms, LNX null mice\",\n      \"journal\": \"The European journal of neuroscience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — reciprocal co-IP plus pull-down plus functional ligase activity comparison in cells, with KO mouse controls\",\n      \"pmids\": [\"30295974\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2018,\n      \"finding\": \"Three unique cysteine residues (C264 in TM4, C92 and C87 in TM2) form a specific docking pocket in Cx36 accessible only to short-chain n-alcohols and isoflurane (which stimulate Cx36 GJ conductance) but not to inhibitory compounds; cysteine substitution mutations at these sites reversed the stimulatory effect of hexanol and isoflurane to inhibitory, suggesting that stimulation involves reshuffling of inter-subunit disulfide bonds (C264-C92) to intra-subunit bonds (C264-C87).\",\n      \"method\": \"Structural modeling/docking, site-directed mutagenesis of cysteines, dual whole-cell patch-clamp in HeLa and N2A cells\",\n      \"journal\": \"Bioscience reports\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 1-2 — mutagenesis with functional electrophysiology, but mechanistic model is partially computational\",\n      \"pmids\": [\"29298877\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2020,\n      \"finding\": \"A Cx36-GCaMP fusion biosensor formed functional gap junctions with PKA-regulated Neurobiotin coupling; local Ca2+ increases around Cx36 gap junctions were resolved in response to Ca2+ ionophore and glutamate application; glutamate potentiated Cx36-GCaMP coupling in HeLa cells via endogenous NMDA/kainate-type glutamate receptors and a Ca2+/CaMKII-dependent mechanism, demonstrating that glutamate receptor activation locally regulates Cx36 coupling strength through Ca2+ and CaMKII.\",\n      \"method\": \"Cx36-GCaMP fusion protein, tracer coupling assay, Ca2+ imaging, pharmacological CaMKII inhibition, RNA-seq confirmation of glutamate receptor expression\",\n      \"journal\": \"eNeuro\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — novel biosensor approach with pharmacological validation; single lab\",\n      \"pmids\": [\"32179580\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2021,\n      \"finding\": \"Depletion of zebrafish gjd2a (Cx35.5, ortholog of mammalian GJD2) caused hyperopia and electrophysiological changes in the retina, while depletion of gjd2b (Cx35.1) caused nuclear cataract that triggered axial elongation; establishing that GJD2 orthologs regulate ocular biometry and refractive status through retinal and lenticular mechanisms.\",\n      \"method\": \"Zebrafish gjd2 morpholino/CRISPR knockouts, ocular biometry, electroretinography, immunohistochemistry, scRNA sequencing\",\n      \"journal\": \"Communications biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — loss-of-function with multiple functional readouts in a vertebrate model, two paralogs distinguished\",\n      \"pmids\": [\"34083742\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2021,\n      \"finding\": \"Cx36 expression in zebrafish ortholog GJD2 and retinal gap junctions (particularly in the IPL) regulate emmetropization; form-deprivation decreased total Cx36 and phosphorylated Cx36 (P-Cx36) and the P-Cx36/Cx36 ratio in the IPL, while pharmacological uncoupling with 18-β-GA induced myopia and reduced Cx36 and P-Cx36 in both IPL and OPL, implicating Cx36 phosphorylation state as a functional readout in refractive development.\",\n      \"method\": \"Form-deprivation and lens-induced myopia in guinea pigs, subconjunctival injection of gap junction uncoupler 18-β-GA, quantitative immunofluorescence of Cx36 and P-Cx36, neurobiotin cut-loading\",\n      \"journal\": \"Investigative ophthalmology & visual science\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — pharmacological and deprivation models with quantitative phospho-Cx36 readout; single lab\",\n      \"pmids\": [\"34283211\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2021,\n      \"finding\": \"CaM and CaMKII binding to Cx36 are both calcium-dependent, with overlapping binding sites on the Cx36 C-terminus; CaM can engage Cx36 outside the gap junction plaque. The review synthesizes evidence that these Ca2+-dependent interactions constitute a core regulatory mechanism for Cx36 electrical synapse plasticity: Ca2+ elevation leads to CaM binding, which activates CaMKII, which phosphorylates Cx36 to modulate coupling strength.\",\n      \"method\": \"Review synthesizing binding assays, NMR structure, electrophysiology, and phosphorylation data from multiple studies\",\n      \"journal\": \"International journal of molecular sciences\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — mechanistic model supported by primary data from multiple prior studies\",\n      \"pmids\": [\"33922931\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2023,\n      \"finding\": \"Cryo-EM structures of human Cx36 gap junction channel at 2.2–3.6 Å resolution revealed a dynamic equilibrium between closed and open states: in the closed state, channel pores are obstructed by lipids with N-terminal helices (NTHs) excluded from the pore; in the open state, pore-lining NTHs create a more acidic pore than Cx26 or Cx46/50, explaining Cx36's strong cation selectivity. Channel opening involves an α-to-π-helix transition in TM1 that weakens protomer-protomer interactions.\",\n      \"method\": \"Cryo-electron microscopy at 2.2–3.6 Å resolution, structural analysis of open and closed states\",\n      \"journal\": \"Nature communications\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — near-atomic resolution cryo-EM structure providing mechanistic basis for cation selectivity and gating\",\n      \"pmids\": [\"36906653\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"Cryo-EM structures of human Cx36 in complex with inhibitors (mefloquine, arachidonic acid, 1-hexanol) showed that all three competitively bind a pocket at the N-terminal helices (NTH), inducing a conformational shift from the pore-lining NTH (PLN) state to a flexible NTH (FN) state that allows lipids to obstruct the channel pore, revealing the molecular mechanism of Cx36 channel inhibition.\",\n      \"method\": \"Cryo-electron microscopy structures with inhibitor-bound Cx36, structural comparison of PLN vs. FN conformational states\",\n      \"journal\": \"Nature communications\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 — cryo-EM structures with three distinct inhibitors revealing a shared competitive binding mechanism\",\n      \"pmids\": [\"39455592\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"The C-terminal tip of Cx36 mediates two distinct interactions in the early secretory pathway: the terminal valine acts as an ER export signal recruiting COPII cargo receptors Sec24A/B/C at ER exit sites, while the PDZ-binding motif SAYV mediates interaction with Golgi-stacking protein Grasp55 to stabilize Cx36 in the Golgi. Sec24 promotes ER exit, while Grasp55 retains Cx36 in the Golgi, establishing opposing regulatory roles in Cx36 trafficking.\",\n      \"method\": \"HEK293T expression system, siRNA knockdown, BioID proximity labeling screens, co-immunoprecipitation, confocal microscopy of ER vs. Golgi localization, overexpression studies\",\n      \"journal\": \"Cellular and molecular life sciences\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — BioID screens combined with functional siRNA and co-IP validation of two distinct trafficking interactions\",\n      \"pmids\": [\"39395036\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"Hemichannel permeability assays showed that Cx36 hemichannels selectively allow ATP release but not glutamate or lactate; mutational analysis of N-terminus charge and N-terminus–TM2 interactions (informed by differential selectivity between Cx46 and Cx50) identified these structural elements as key determinants of permselectivity, showing that molecular size alone cannot explain connexin hemichannel selectivity.\",\n      \"method\": \"Co-expression of genetically encoded fluorescent sensors for ATP, glutamate, and lactate with connexins in cells; N-terminus mutagenesis; pharmacological stimulation\",\n      \"journal\": \"bioRxiv (preprint)\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 1-2 — functional permeability assay with mutagenesis, but preprint only\",\n      \"pmids\": [\"bio_10.1101_2025.03.12.642803\"],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"Electrostatic interactions (ESIs) at the second extracellular (E2) loop interface are required for Cx36 gap junction formation: at least three ESI residue pairs per E2-E2 interface must be intact for functional GJ conductance. These unique ESIs also ensure Cx36 docking specificity to itself, preventing heterotypic GJ formation with other brain connexins (Cx26, Cx30, Cx31.3, Cx32, Cx43, Cx45, Cx47).\",\n      \"method\": \"Computational electrostatic calculations, systematic missense mutagenesis of E2 interface residues, dual patch-clamp in HEK293 cell pairs\",\n      \"journal\": \"bioRxiv (preprint)\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 1-2 — structure-guided mutagenesis with electrophysiological validation, but preprint only\",\n      \"pmids\": [\"bio_10.1101_2025.10.25.684567\"],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"ZO-1 interacts with Cx36 via PDZ1 domain binding to the Cx36 PDZ-binding motif (SAYV), and this interaction is inherently weak; molecular dynamics simulations and binding assays identified a single substitution at position 319 and acidic residues adjacent to the PBM that evolutionarily weaken this interaction. CaMKII-mediated phosphorylation of Cx36 reduces ZO-1 binding, suggesting that ZO-1 unbinding is a necessary event during potentiation of electrical synapses.\",\n      \"method\": \"Gaussian accelerated molecular dynamics, binding affinity assays, site-directed mutagenesis at PBM and adjacent residues, CaMKII phosphorylation assays\",\n      \"journal\": \"bioRxiv (preprint)\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 1-2 — computational + binding assay + phosphorylation data, but preprint only\",\n      \"pmids\": [\"bio_10.1101_2025.10.27.684941\"],\n      \"is_preprint\": true\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"Cx36 gap junctions in the outer retina lower visual thresholds (i.e., increase sensitivity) under dim light, while inner retinal Cx36 gap junctions are required for maintaining thresholds; complete Cx36 KO increased visual thresholds 16.5-fold relative to controls, and outer retina-specific Cx36 removal lowered thresholds 2.6-fold, demonstrating layer-specific and opposing roles of Cx36 in setting visual detection thresholds via the rod OFF-pathway.\",\n      \"method\": \"Transgenic mice with retinal layer-specific Cx36 knockout, operant behavioral assay, psychophysical modeling\",\n      \"journal\": \"iScience\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — layer-specific genetic knockout with quantitative behavioral readout and computational modeling\",\n      \"pmids\": [\"41623460\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2025,\n      \"finding\": \"In pancreatic islets, delta cell activation precedes beta cell Ca2+ oscillations at high glucose, and selective Cx36 knockout in delta cells confirmed that low-density Cx36 gap junctions contribute to delta-beta cell coordination; however, blockade of vesicle release (paracrine signaling) eliminated coupling between most delta and beta cells, demonstrating that beta-delta coordination is primarily driven by paracrine signals (Urocortin 3) with secondary contribution from Cx36 gap junctions.\",\n      \"method\": \"GCaMP6s Ca2+ imaging of hundreds of cells simultaneously, selective delta-cell Cx36 knockout, Rho-GTPase inhibitor ML-141 to block vesicle release\",\n      \"journal\": \"Proceedings of the National Academy of Sciences of the United States of America\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 — cell-type-specific KO combined with pharmacological dissection and large-scale Ca2+ imaging\",\n      \"pmids\": [\"40956879\"],\n      \"is_preprint\": false\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"In vivo BioID proximity labeling using Cx36-EGFP in AII amacrine cells identified more than 50 new proteins associated with Cx36 electrical synapses in mouse retina, including scaffold proteins, adhesion molecules, and cytoskeletal regulators; ZO-1 and ZO-2 were confirmed, and Sipa1l3 was identified as a new electrical synapse scaffold that interacts with ZO-1, ZO-2, and Cx36.\",\n      \"method\": \"In vivo BioID proximity labeling (GFP-nanobody-TurboID in mice; Cx35b-TurboID transgenic zebrafish), mass spectrometry, immunofluorescence, binding interaction assays\",\n      \"journal\": \"bioRxiv (preprint)\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 — in vivo proximity labeling with cross-species validation and direct binding confirmation for Sipa1l3, but preprint only\",\n      \"pmids\": [\"bio_10.1101_2024.11.26.625481\"],\n      \"is_preprint\": true\n    }\n  ],\n  \"current_model\": \"GJD2-encoded Cx36 is the principal gap junction protein of neuronal electrical synapses and pancreatic beta-cells: it forms homotypic, strongly cation-selective gap junction channels whose open/closed states involve lipid occlusion and N-terminal helix repositioning (established by cryo-EM); channel conductance is potentiated by Ca²⁺-loaded calmodulin binding to the C-terminal tail (W277/V284 anchors) and by CaMKII phosphorylation of the same domain (analogous to NR2B regulation), while LNX1/LNX2 E3 ubiquitin ligases ubiquitinate Cx36 to drive its internalization; trafficking from ER to Golgi is controlled by C-terminal valine (COPII/Sec24 ER export signal) and the SAYV PDZ-binding motif (Grasp55 Golgi retention), with ZO-1/ZO-2/ZO-3 anchoring Cx36 at synaptic plaques via PDZ1 interactions that are weakened by CaMKII phosphorylation; in beta-cells, Cx36-dependent coupling synchronizes Ca²⁺ oscillations and insulin secretion, and its expression is transcriptionally controlled by Beta2/NeuroD1 (activator) and NRSF/REST (repressor); in the retina, Cx36 in distinct layers differentially controls visual thresholds and emmetropization.\"\n}\n```","stage2_raw":"```json\n{\n  \"mechanistic_narrative\": \"GJD2 encodes connexin-36 (Cx36), the principal gap junction protein of neurons and pancreatic beta-cells, where it assembles into electrical synapses and intercellular channels that synchronize network activity, coordinate calcium oscillations, and regulate insulin secretion [PMID:9753189, PMID:10905480, PMID:17395748]. Cx36 channel gating is controlled by N-terminal helix conformational transitions—inhibitors such as mefloquine, arachidonic acid, and alcohols bind the N-terminal helix pocket and shift it from a pore-lining to a flexible state that occludes the pore—while unique TM2/TM4 cysteine residues mediate stimulatory modulation by short-chain alcohols through inter-subunit disulfide rearrangement [PMID:39455592, PMID:29298877]. Ca²⁺-loaded calmodulin binds the Cx36 C-terminus to activate channels, competing with CaMKII for an overlapping site; CaMKII-dependent phosphorylation at Ser293 further tunes coupling strength, and the C-terminal PDZ-binding motif SAYV mediates both ZO-1 anchoring at electrical synapses and Grasp55-dependent Golgi stabilization during COPII/Sec24-mediated ER export [PMID:27917108, PMID:39395036]. In beta-cells, Cx36 gap junctions suppress excitability of coupled cells, coordinate stimulus-secretion coupling, and integrate paracrine signals from delta cells; loss of Cx36 alters calcium dynamics and insulin release, and a synonymous GJD2 polymorphism (rs3743123) impairs plaque formation and causes hyperglycemia in transgenic mice [PMID:24458355, PMID:26959991, PMID:40956879].\",\n  \"teleology\": [\n    {\n      \"year\": 1998,\n      \"claim\": \"Identification of Cx36 as the first predominantly neuronal connexin in mammals resolved the long-standing question of which gap junction protein mediates electrical synapses in the CNS.\",\n      \"evidence\": \"Degenerate RT-PCR cloning, in situ hybridization, and neurotoxin lesion experiments in rodent brain\",\n      \"pmids\": [\"9753189\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Channel properties (conductance, selectivity) not yet characterized\", \"No loss-of-function data at this stage\"]\n    },\n    {\n      \"year\": 2000,\n      \"claim\": \"Demonstration that Cx36 is selectively expressed by beta-cells within pancreatic islets established a non-neuronal role and identified the connexin responsible for beta-cell electrical coupling.\",\n      \"evidence\": \"In situ hybridization and immunolabeling of FACS-purified beta vs. non-beta cell preparations\",\n      \"pmids\": [\"10905480\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Functional consequence of Cx36 loss in beta-cells not yet tested\", \"Whether Cx36 forms heterotypic junctions with other islet cell types unknown\"]\n    },\n    {\n      \"year\": 2001,\n      \"claim\": \"Ultrastructural mapping confirmed Cx36 exclusivity to neuronal gap junctions while other connexins segregated to glia, establishing that distinct connexin-defined communication networks coexist in the CNS.\",\n      \"evidence\": \"Freeze-fracture replica immunogold labeling (FRIL) of >4000 junctions in adult rat CNS\",\n      \"pmids\": [\"12064610\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Cell-type specificity among neuronal subtypes not resolved\", \"Developmental dynamics of Cx36 expression not addressed\"]\n    },\n    {\n      \"year\": 2001,\n      \"claim\": \"Cx36 knockout mice and antisense knockdown revealed that Cx36 is dispensable for beta-cell development but required for normal insulin secretion, separating structural from functional roles.\",\n      \"evidence\": \"Cx36 KO mice and antisense transfection in MIN6 cells with insulin secretion assays\",\n      \"pmids\": [\"12064624\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Mechanism linking Cx36 loss to secretory defect (Ca²⁺ dynamics vs. metabolic coupling) unresolved\", \"In vivo glucose homeostasis phenotype not fully characterized\"]\n    },\n    {\n      \"year\": 2002,\n      \"claim\": \"Mutagenesis of the N-terminal domain established its requirement for proper Cx36 trafficking, while ruling out phosphorylation sites as determinants of subcellular localization.\",\n      \"evidence\": \"Site-directed mutagenesis of Cx36-EGFP in neuroblastoma cells and hippocampal neurons with electrophysiology\",\n      \"pmids\": [\"12210839\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"Single-lab study; N-terminal trafficking signals not mapped to specific residues\", \"Interaction partners mediating N-terminal-dependent trafficking unknown\"]\n    },\n    {\n      \"year\": 2004,\n      \"claim\": \"Pharmacological profiling identified mefloquine as a potent and selective Cx36 channel blocker, providing the first connexin-subtype-selective tool compound for studying electrical synapses.\",\n      \"evidence\": \"Dual patch-clamp in N2A cells expressing defined connexins plus validation in neocortical slices\",\n      \"pmids\": [\"15297615\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Binding site on Cx36 unknown at this stage\", \"Off-target effects of mefloquine in vivo not fully excluded\"]\n    },\n    {\n      \"year\": 2007,\n      \"claim\": \"Electrophysiology in pancreatic slices from Cx36-null mice revealed that Cx36-mediated coupling hyperpolarizes active beta-cells through inactive neighbors, defining the electrical mechanism by which gap junctions set excitability thresholds and shape insulin release kinetics.\",\n      \"evidence\": \"Patch-clamp recordings in pancreatic slices from Cx36−/−, Cx36+/−, and Cx36+/+ mice\",\n      \"pmids\": [\"17395748\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"How coupling heterogeneity across the islet modulates pulsatile secretion not resolved\", \"Contribution of metabolic vs. electrical coupling through Cx36 not separated\"]\n    },\n    {\n      \"year\": 2014,\n      \"claim\": \"Genetic epistasis showed that Cx36 knockout rescues glucose-stimulated Ca²⁺ activity and insulin secretion in mice with ATP-insensitive KATP channels, proving Cx36 coupling suppresses beta-cell excitability and can be a therapeutic target for KATP gain-of-function diabetes.\",\n      \"evidence\": \"Double-knockout mouse model with Ca²⁺ imaging, insulin secretion assays, and glucose tolerance tests\",\n      \"pmids\": [\"24458355\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Long-term metabolic consequences of Cx36 loss in this model not assessed\", \"Whether partial coupling reduction achieves similar rescue unknown\"]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"NMR solution structure of the CaM–Cx36 C-terminal peptide complex revealed Ca²⁺-dependent CaM binding that activates Cx36 channels and competes with CaMKII, establishing a dual Ca²⁺-sensing mechanism unique among connexins.\",\n      \"evidence\": \"NMR structure determination, co-immunoprecipitation, electrophysiology, Ca²⁺-dependence assays\",\n      \"pmids\": [\"27917108\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Full-length structural context of CaM binding unknown\", \"How CaM vs. CaMKII competition is regulated in vivo not determined\"]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"A synonymous GJD2 SNP (rs3743123) was shown to impair gap junction plaque formation, reduce islet Cx36 levels, and cause hyperglycemia in transgenic mice, demonstrating that silent coding variants can functionally disrupt Cx36-dependent beta-cell coupling.\",\n      \"evidence\": \"HeLa cell transfection, plaque imaging, coupling assays, and transgenic mouse lines with insulin-promoter expression\",\n      \"pmids\": [\"26959991\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Molecular mechanism by which the synonymous change reduces protein levels (mRNA stability vs. translation) not determined\", \"Human population-level metabolic impact not established\"]\n    },\n    {\n      \"year\": 2018,\n      \"claim\": \"Identification of three unique TM2/TM4 cysteine residues as the structural basis for alcohol- and anesthetic-induced stimulation of Cx36 channels explained why Cx36 responds oppositely to these agents compared to other connexins.\",\n      \"evidence\": \"Cysteine mutagenesis combined with dual whole-cell patch-clamp in HeLa and N2A cells\",\n      \"pmids\": [\"29298877\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Disulfide rearrangement mechanism inferred from mutagenesis, not directly observed structurally\", \"Physiological relevance of alcohol-mediated stimulation in brain circuits not tested\"]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"Cryo-EM structures of Cx36 in complex with mefloquine, arachidonic acid, and hexanol revealed a common inhibitory gating mechanism: all three bind the N-terminal helix pocket and shift the channel from a pore-lining to a flexible NTH state, allowing lipid bilayer intrusion to occlude the pore.\",\n      \"evidence\": \"Cryo-EM structure determination of purified human Cx36 gap junction channels with inhibitors\",\n      \"pmids\": [\"39455592\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Open-state high-resolution structure not obtained\", \"How CaM-mediated activation interfaces with NTH conformational states structurally unresolved\"]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"BioID proximity labeling and siRNA experiments mapped the Cx36 biosynthetic pathway: the C-terminal valine acts as a COPII/Sec24A/B/C ER export signal, while the SAYV PDZ-binding motif recruits Grasp55 for Golgi stabilization, defining the trafficking machinery for electrical synapse assembly.\",\n      \"evidence\": \"BioID screens and siRNA knockdowns in HEK293T cells with confocal imaging\",\n      \"pmids\": [\"39395036\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Whether this trafficking pathway operates identically in neurons and beta-cells not tested\", \"Post-Golgi sorting to gap junction plaques not characterized\"]\n    },\n    {\n      \"year\": 2025,\n      \"claim\": \"Cell-type-specific Cx36 knockout in delta cells combined with live Ca²⁺ imaging revealed that delta-beta cell coordination uses both low-density Cx36 gap junctions and paracrine vesicle release, integrating electrical and chemical signaling modes within islets.\",\n      \"evidence\": \"GCaMP6s Ca²⁺ imaging of hundreds of cells, delta-cell-specific Cx36 KO, pharmacological blockade of vesicle release\",\n      \"pmids\": [\"40956879\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Identity of paracrine factor(s) mediating vesicle-dependent coupling not determined\", \"How delta-beta Cx36 junctions form given Cx36 homospecificity constraints unknown\"]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"Key unresolved questions include the high-resolution open-state structure of Cx36, the mechanism by which CaM-mediated activation and CaMKII phosphorylation cooperatively regulate NTH gating conformations, and how Cx36 trafficking and plaque dynamics are controlled at neuronal electrical synapses in vivo.\",\n      \"evidence\": \"\",\n      \"pmids\": [],\n      \"confidence\": \"High\",\n      \"gaps\": [\"No open-state cryo-EM structure available\", \"In vivo dynamics of Cx36 plaque assembly/disassembly at electrical synapses unresolved\", \"Structural basis for CaM activation vs. CaMKII potentiation not integrated into a single model\"]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0005215\", \"supporting_discovery_ids\": [0, 8, 9, 10, 17, 21, 25]},\n      {\"term_id\": \"GO:0005198\", \"supporting_discovery_ids\": [3, 20, 22]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005886\", \"supporting_discovery_ids\": [3, 10, 20]},\n      {\"term_id\": \"GO:0005783\", \"supporting_discovery_ids\": [5, 20]},\n      {\"term_id\": \"GO:0005794\", \"supporting_discovery_ids\": [20]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"R-HSA-112316\", \"supporting_discovery_ids\": [0, 3, 8, 9, 18, 26, 27]},\n      {\"term_id\": \"R-HSA-1500931\", \"supporting_discovery_ids\": [9, 10, 22, 28]},\n      {\"term_id\": \"R-HSA-9609507\", \"supporting_discovery_ids\": [5, 20]},\n      {\"term_id\": \"R-HSA-162582\", \"supporting_discovery_ids\": [15, 18]}\n    ],\n    \"complexes\": [\n      \"Cx36 homotypic gap junction channel\"\n    ],\n    \"partners\": [\n      \"CALM1\",\n      \"CAMK2B\",\n      \"CAMK2D\",\n      \"TJP1\",\n      \"SEC24A\",\n      \"GORASP2\",\n      \"SIPA1L3\"\n    ],\n    \"other_free_text\": []\n  }\n}\n```\n\n--- ROUND 2 (R2/R3, augmented corpus) ---\n\n```json\n{\n  \"mechanistic_narrative\": \"GJD2 encodes connexin 36 (Cx36), the principal gap junction protein of neuronal electrical synapses and pancreatic beta-cells, where it forms homotypic, strongly cation-selective intercellular channels that synchronize membrane potential and Ca²⁺ oscillations [PMID:9753189, PMID:10905480, PMID:22752717]. Cryo-EM structures reveal that channel gating involves an α-to-π-helix transition in TM1 that repositions N-terminal helices into the pore lumen, while channel closure is achieved by lipid occlusion when inhibitors (mefloquine, arachidonic acid) competitively displace N-terminal helices from their pore-lining configuration [PMID:36906653, PMID:39455592]. Channel conductance is potentiated by Ca²⁺/calmodulin binding at the C-terminal tail (W277/V284 anchors) and by CaMKII phosphorylation of overlapping cytoplasmic domains, while CaMKII phosphorylation simultaneously weakens ZO-1 scaffold interactions to remodel electrical synapse architecture; LNX1/LNX2 E3 ubiquitin ligases ubiquitinate Cx36 to drive gap junction plaque internalization [PMID:27917108, PMID:19095792, PMID:19418635, PMID:30295974]. In beta-cells, Cx36-dependent coupling propagates hyperpolarizing currents to coordinate insulin secretion, with transcription controlled by Beta2/NeuroD1 (activator) and NRSF/REST (repressor), while in the retina, layer-specific Cx36 gap junctions differentially set visual detection thresholds and regulate emmetropization [PMID:17395748, PMID:22729650, PMID:14565956, PMID:41623460, PMID:34083742].\",\n  \"teleology\": [\n    {\n      \"year\": 1998,\n      \"claim\": \"The identity of a neuron-specific connexin was unknown; cloning of Cx36 from mammalian brain established GJD2 as the first connexin expressed preferentially in neurons rather than glia, defining a new molecular basis for neuronal electrical synapses.\",\n      \"evidence\": \"Degenerate RT-PCR cloning, in situ hybridization with neurotoxin lesion validation in rat brain\",\n      \"pmids\": [\"9753189\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Functional channel properties not yet determined\", \"No loss-of-function data\"]\n    },\n    {\n      \"year\": 2000,\n      \"claim\": \"Whether Cx36 operated outside the CNS was unclear; detection of Cx36 selectively in pancreatic beta-cells established a second major physiological context and linked gap junctions to insulin secretion biology.\",\n      \"evidence\": \"In situ hybridization, immunolabeling, FACS-purified beta- vs. non-beta-cell fractions in rodent islets\",\n      \"pmids\": [\"10905480\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Functional consequence for insulin secretion not yet demonstrated\", \"Human beta-cell expression not confirmed at this point\"]\n    },\n    {\n      \"year\": 2001,\n      \"claim\": \"Whether neuronal and glial gap junctions share connexins was debated; ultrastructural FRIL demonstrated that Cx36 is exclusively neuronal, with no overlap with astrocytic (Cx43/Cx30/Cx26) or oligodendrocytic (Cx32) connexins.\",\n      \"evidence\": \"Freeze-fracture replica immunogold labeling of >4000 gap junctions in adult rat CNS\",\n      \"pmids\": [\"12064610\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Regional variation in co-expression with other neuronal connexins (e.g. Cx45) not fully resolved\"]\n    },\n    {\n      \"year\": 2003,\n      \"claim\": \"How Cx36 expression is restricted to neurons and beta-cells was unknown; identification of NRSF/REST as a transcriptional repressor acting via an NRSE element in the GJD2 promoter explained tissue specificity—cells lacking NRSF (neurons, beta-cells) express Cx36.\",\n      \"evidence\": \"Luciferase reporter assays, NRSF viral gain-of-function in beta-cell lines, NRSE mutagenesis, HDAC inhibitor treatment\",\n      \"pmids\": [\"14565956\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Other transcription factors (beyond NRSF) controlling Cx36 not yet identified\", \"Chromatin-level regulation not mapped\"]\n    },\n    {\n      \"year\": 2004,\n      \"claim\": \"Whether Cx36 was functionally required for beta-cell secretion (versus a bystander) was unresolved; lentiviral rescue of Cx36 (but not E-cadherin) in Cx36-depleted MIN6 cells demonstrated specific and non-redundant control of insulin secretion by Cx36 gap junctions.\",\n      \"evidence\": \"Stable antisense transfection, lentiviral rescue, insulin secretion assays with multiple secretagogues\",\n      \"pmids\": [\"15023528\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"In vivo relevance in whole-animal glucose homeostasis not yet tested\"]\n    },\n    {\n      \"year\": 2007,\n      \"claim\": \"The biophysical mechanism by which Cx36 controls beta-cell excitability was unclear; electrophysiology in Cx36 KO pancreatic slices showed that Cx36 propagates hyperpolarizing currents from inactive cells, dampening excitability across the islet, and that Cx36 channels are strongly cation-selective.\",\n      \"evidence\": \"Patch-clamp electrophysiology in pancreatic slices from Cx36-/-, +/-, +/+ mice; tracer permeability with multiple charged dyes\",\n      \"pmids\": [\"17395748\", \"17828386\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Structural basis for cation selectivity not yet determined\", \"Molecular identity of permeant species in vivo unknown\"]\n    },\n    {\n      \"year\": 2008,\n      \"claim\": \"How Cx36 coupling strength is dynamically regulated was unknown; discovery that CaMKII directly binds and phosphorylates Cx36 at cytoplasmic domains analogous to the CaMKII–NR2B interaction established a Ca²⁺-dependent plasticity mechanism for electrical synapses.\",\n      \"evidence\": \"In vitro CaMKII binding and phosphorylation assays with Cx36 cytoplasmic peptides, immunofluorescence colocalization in inferior olive\",\n      \"pmids\": [\"19095792\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Phosphorylation sites not mapped to specific residues at this stage\", \"In vivo consequence of phosphorylation on coupling not demonstrated\"]\n    },\n    {\n      \"year\": 2009,\n      \"claim\": \"How Cx36 is anchored at gap junction plaques was unresolved; identification of ZO-2 and ZO-3 as PDZ1-domain partners binding the C-terminal SAYV motif established a scaffold complex at Cx36 electrical synapses.\",\n      \"evidence\": \"Reciprocal co-immunoprecipitation, in vitro PDZ pull-down, C-terminal truncation mutagenesis in HeLa and betaTC-3 cells\",\n      \"pmids\": [\"19418635\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Whether ZO-1 also binds Cx36 directly not confirmed until later studies\", \"Functional consequence of scaffold disruption on coupling not tested\"]\n    },\n    {\n      \"year\": 2012,\n      \"claim\": \"The developmental transcriptional activator of Cx36 in beta-cells was unknown; Beta2/NeuroD1 was shown to bind E-boxes in the GJD2 promoter and transactivate it, establishing Cx36 as a direct transcriptional target during prenatal beta-cell differentiation.\",\n      \"evidence\": \"Reporter gene assays with GJD2 promoter fragments, E-box binding assays, lentiviral NeuroD1 expression, developmental immunolabeling\",\n      \"pmids\": [\"22729650\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Whether NeuroD1 also controls Cx36 in neurons not tested\", \"Epigenetic regulation of the GJD2 locus not addressed\"]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"How Ca²⁺ directly modulates Cx36 channels was structurally undefined; NMR resolution of the CaM–Cx36 C-terminal complex revealed Ca²⁺-loaded CaM binds at W277/V284 in a compact conformation and activates Cx36 channels, unlike its inhibitory role on other connexins.\",\n      \"evidence\": \"NMR solution structure, Ca²⁺-dependent binding assays, electrophysiological channel activation measurements\",\n      \"pmids\": [\"27917108\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Full-length Cx36–CaM complex structure not available\", \"Whether CaM and CaMKII binding are mutually exclusive in situ remains unclear\"]\n    },\n    {\n      \"year\": 2018,\n      \"claim\": \"How Cx36 gap junctions are turned over was unknown; LNX1/LNX2 E3 ubiquitin ligases were identified as direct interactors that ubiquitinate Cx36 and drive gap junction plaque internalization, establishing ubiquitin-dependent degradation as a removal mechanism.\",\n      \"evidence\": \"Co-IP, PDZ pull-down, cotransfection of ligase-active vs. ligase-dead LNX isoforms, LNX null mice\",\n      \"pmids\": [\"30295974\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Specific ubiquitination sites on Cx36 not mapped\", \"Whether LNX acts at all Cx36-expressing tissues not tested\"]\n    },\n    {\n      \"year\": 2023,\n      \"claim\": \"The structural basis for Cx36 gating and cation selectivity was unknown; cryo-EM structures at 2.2–3.6 Å revealed that channel opening requires an α-to-π-helix transition in TM1 that positions N-terminal helices into a negatively charged pore, while closure involves lipid occlusion of the channel.\",\n      \"evidence\": \"Cryo-electron microscopy of human Cx36 gap junction channels in open and closed conformations\",\n      \"pmids\": [\"36906653\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Dynamics of lipid entry/exit not resolved\", \"No structures with CaM or CaMKII-phosphorylated Cx36\"]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"How pharmacological inhibitors block Cx36 was structurally undefined; cryo-EM structures with mefloquine, arachidonic acid, and hexanol showed all three bind a common N-terminal helix pocket, shifting it from a pore-lining to a flexible state that permits lipid occlusion, unifying the inhibition mechanism.\",\n      \"evidence\": \"Cryo-EM structures of inhibitor-bound Cx36 with conformational state analysis\",\n      \"pmids\": [\"39455592\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"No activator-bound structures available\", \"Whether this mechanism extends to in vivo neuromodulatory inhibition is untested\"]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"How Cx36 traffics from ER to plasma membrane was poorly understood; identification of the C-terminal valine as a COPII/Sec24 ER-export signal and the SAYV PDZ motif as a Grasp55-mediated Golgi retention signal established opposing regulatory checkpoints in the early secretory pathway.\",\n      \"evidence\": \"BioID proximity labeling, siRNA knockdown, co-immunoprecipitation, confocal ER/Golgi localization in HEK293T cells\",\n      \"pmids\": [\"39395036\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Whether these trafficking signals are regulated by phosphorylation in vivo is unknown\", \"Post-Golgi trafficking steps not mapped\"]\n    },\n    {\n      \"year\": 2025,\n      \"claim\": \"Whether Cx36 gap junctions serve identical functions across retinal layers was unclear; layer-specific knockout showed that outer retinal Cx36 lowers visual thresholds (increases sensitivity) while inner retinal Cx36 maintains thresholds, revealing opposing functional roles of the same connexin in different circuit contexts.\",\n      \"evidence\": \"Transgenic mice with retinal layer-specific Cx36 knockout, operant behavioral assay, psychophysical modeling\",\n      \"pmids\": [\"41623460\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Circuit-level mechanism by which inner vs. outer Cx36 differentially set thresholds not fully resolved\", \"Contribution of Cx36 phosphorylation state in each layer not tested\"]\n    },\n    {\n      \"year\": 2025,\n      \"claim\": \"The relative contribution of Cx36 gap junctions versus paracrine signaling to delta-beta cell coordination in islets was debated; selective delta-cell Cx36 KO and vesicle-release blockade demonstrated that paracrine signals (Urocortin 3) dominate, with Cx36 providing a secondary coordination mechanism.\",\n      \"evidence\": \"GCaMP6s Ca²⁺ imaging, delta-cell-specific Cx36 knockout, Rho-GTPase inhibitor to block vesicle release\",\n      \"pmids\": [\"40956879\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"Whether Cx36 coupling between delta and beta cells is physiologically regulated remains unclear\", \"Direct measurement of delta-beta electrical coupling not performed\"]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"Key unresolved questions include: (1) the full-length structure of Cx36 in complex with CaM or in a CaMKII-phosphorylated state; (2) how post-Golgi trafficking delivers Cx36 to specific synaptic sites; (3) the identity and roles of most proteins in the Cx36 electrical synapse proteome identified by proximity labeling; and (4) whether Cx36 dysfunction directly causes human Mendelian disease.\",\n      \"evidence\": \"\",\n      \"pmids\": [],\n      \"confidence\": \"Low\",\n      \"gaps\": [\"No CaM-bound or phospho-Cx36 cryo-EM structure\", \"Post-Golgi trafficking pathway undefined\", \"Most BioID-identified interactors lack functional validation\", \"No causative human mutations in GJD2 linked to Mendelian disease with direct evidence\"]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0005215\", \"supporting_discovery_ids\": [0, 8, 14, 19, 32, 33]},\n      {\"term_id\": \"GO:0098631\", \"supporting_discovery_ids\": [3, 12, 32]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005886\", \"supporting_discovery_ids\": [3, 12, 16, 32, 33]},\n      {\"term_id\": \"GO:0005829\", \"supporting_discovery_ids\": [5]},\n      {\"term_id\": \"GO:0005794\", \"supporting_discovery_ids\": [34]},\n      {\"term_id\": \"GO:0005783\", \"supporting_discovery_ids\": [34]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"R-HSA-1500931\", \"supporting_discovery_ids\": [3, 12, 13, 14, 32, 33]},\n      {\"term_id\": \"R-HSA-112316\", \"supporting_discovery_ids\": [0, 3, 8, 12, 15, 38]},\n      {\"term_id\": \"R-HSA-162582\", \"supporting_discovery_ids\": [7, 13, 15, 20, 21, 28]},\n      {\"term_id\": \"R-HSA-9609507\", \"supporting_discovery_ids\": [34]},\n      {\"term_id\": \"R-HSA-382551\", \"supporting_discovery_ids\": [14, 19, 32]}\n    ],\n    \"complexes\": [\n      \"Cx36 homotypic gap junction channel (dodecameric)\"\n    ],\n    \"partners\": [\n      \"TJP1\",\n      \"TJP2\",\n      \"TJP3\",\n      \"CALM1\",\n      \"CAMK2B\",\n      \"LNX1\",\n      \"LNX2\",\n      \"GORASP2\"\n    ],\n    \"other_free_text\": []\n  }\n}\n```"}